PodcastsHear the voice. See the shape of the thought.

Fiona Fung, who runs engineering and product for Claude Code and Cowie at Anthropic, walks through what broke when agentic coding became the team's default — review, ownership, planning, hiring — and the norms they rewrote to keep shipping. The throughline: when coding stops being the bottleneck, every process built around protecting expensive engineering bandwidth quietly stops working, and the manager's job is to notice and rewrite them fast. ## [00:00] Intro and the five themes Fiona opens with a confession that the room is much fuller than she expected (Boris and Jared's session is still letting out), takes a selfie with the audience, and frames the talk. Background: she grew teams at Meta and Microsoft before Anthropic, and is now responsible for Claude Code and Cowie engineering and product. The deck she's about to walk through has already been rewritten in the past month — routines didn't exist when she first wrote the slides. She previews five threads: bottlenecks have shifted, team norms had to be rewritten, how they rolled them out, what signals say the changes are working, and the open questions she's still sitting with. > *"I did this slide deck maybe like a month ago and already I've had to change some of the content cuz when I started this deck, there were no routines."* ## [02:10] The shift: bottlenecks have moved Fiona's subtitle for the whole talk is *what served you prior may not serve you any longer*. She takes the audience back to shipping Visual Studio 2005 on CD-ROMs — hard deadlines because the manufacturing lab had to print discs — and points out that the move from CDs to online distribution already rewired how teams ship. The new shift is bigger: for years coding throughput and engineering bandwidth were the expensive things, and that's quietly stopped being true on Claude Code. When the bottleneck moves, it doesn't disappear — it relocates to verification, review, cross-functional handoffs, and security. The questions that matter now are "is this code correct?" and "is this safe?", and the old planning and ownership norms quietly stop serving the team. > *"What served you prior may not serve you any longer."* ## [07:40] Rewriting team norms: code review, JIT planning, technical debates Inside Claude Code the team had to rewrite the norms one by one. Code review is the first — human judgment shifts to "who actually needs to look at this." Planning is the second — Fiona calls it JIT planning, like JIT compiling, because prototyping is no longer the expensive step that justifies a six-month roadmap. Technical debates are the third: code wins. Instead of two engineers arguing on a doc, both prototype the API and look at impact on callers, and Fiona made a point of caring about the API's downstream effects as much as the implementation itself. The unifying rule: when building is cheap and arguing is expensive, you don't let the last person who checks in win — you build the routines that get *you* the last word. > *"When building is cheap, arguing expensive, again, how does that shift your team norms a bit?"* ## [13:30] Routines and Claude as a second pair of hands With morning coffee Fiona now reads what a routine produced overnight rather than kicking off the work herself. The team leans on Claude code review heavily — Claude babysits PRs, handles styling, lint, and feedback requests, catches bugs before commit, and adds tests — while humans focus on the calls where trust is still being built. She also stresses product sense in tooling: she themed Claude's terminal output ice blue with snowflakes over the holidays, then pulls back to the bigger point that catching bugs earlier (shift left) and automating the double-click question matter more than any one tool. > *"Where do you trust Claude a lot, but then where do you still want a human?"* ## [16:45] Cross-functional gaps and hiring for the hard parts Fiona walks through a survey-update story: she didn't have a dedicated content designer, so Claude became her partner for terse, terminal-appropriate copy. Meanwhile PMs on the team write code, and engineers lean into PM work. The flip-side conclusion for hiring: non-traditional coders can now do more engineering, so the leader's job is to double down on the hard parts the team is actually missing. When she joined, Claude Code was strong on product generalists and creative folks but thin on distributed-systems expertise — that's where she pushed recruiting. > *"With Claude, you have non-traditional coders now being able to do more engineering, but you also have engineers that we can also now lean in to do other roles."* ## [18:51] Flat org and answering customer feedback yourself Fiona pushed her recruiters into an uncomfortable place: hire managers, but have them start as ICs first. The recruiter thought she was crazy; Fiona's answer is that dogfooding Claude Code is the job, and if a candidate isn't up for it the team is better off finding out early. Flat structure plus Claude as a context-switching aid is what lets her, as a manager, still ship code and answer customer requests directly from her desktop Claude Code — instead of routing every customer question through a triage system, she pulls up the local repository and answers it herself. > *"You want to hire managers and they will start as an IC first. No manager would be interested in that."* ## [25:00] Signals you're trending right and open questions The team's working metric is unglamorous and direct: every commit is cloud-assisted by default, and Fiona hasn't seen a non-Claude commit in roughly four months. But she warns against fetishizing the "X percent of code generated by AI" headline — throughput is one signal, not the goal. The end question is what product you're making more delightful and what problem you're solving, with quality and reliability watched alongside volume. She closes with the section she calls "audit your own effort," opens up the questions she's still asking herself, and hands suggestions back to the audience to take to their own teams. > *"For us, it's by default every commit is cloud-assisted. I don't think I've seen a non-cloud-assisted commit probably in the last 4 months or so."* ## Entities - **Fiona Fung** (Person): Director of Engineering at Anthropic, runs Claude Code and Cowie engineering + product; previously led teams at Meta and Microsoft. - **Boris** (Person): Engineering lead on Claude Code, frequent collaborator referenced throughout. - **Kat (Cat)** (Person): Anthropic colleague who gave a keynote earlier the same day on Claude code review. - **Claude Code** (Software): Anthropic's agentic coding tool that is now the default for the team Fiona runs. - **Cowie** (Software): Sister product Fiona's team also owns engineering + product for. - **Anthropic** (Organization): The company building Claude and Claude Code. - **JIT planning** (Concept): Fiona's term for shifting from a six-month roadmap to just-in-time planning, modeled on JIT compilation. - **Shift left** (Concept): Moving bug-catching and verification earlier — into automation and tooling — instead of relying on review after the fact. - **Routines** (Concept): Repeatable Claude-driven workflows the team relies on so a single human gets the last word on outcomes rather than the last commit timestamp winning.

Ben Horowitz and David Ulevitch — recorded at a16z's American Dynamism Summit in Washington — cover the full arc of what it means for a venture firm to accept industry leadership: from America's race to integrate AI into national defense, to the real reason the Anthropic–Department of War deal collapsed, to why the VC industry is consolidating around large generalist firms and narrow specialists. Horowitz closes on what he sees as America's most underrated strategic risk: a profound pessimism about AI at home while China and Japan charge forward with optimism. ## [00:00] Trailer The opening montage frames the episode's central tension: over 70% of Chinese citizens are optimistic about AI, while fewer than 30% of Americans share that view. David Ulevitch sets the stakes — a16z has placed the largest venture bet in American history on the proposition that the U.S. will win the next century of technology. > *"Over 70% of people in China are optimistic about AI and less than 30% in America were optimistic about AI."* ## [00:41] Why America's Technology Dominance Matters for the World Following a16z's record $15 billion fundraise — the largest in the firm's history — David Ulevitch asks what obligations accompany that scale. Horowitz reaches back to advice from his mentor Andy Grove: when you lead an industry, the entire industry's ethics and morality depends on you. He translates that into a first-principles argument: what matters for humanity is whether people have a genuine chance to contribute, and no country comes close to America on that dimension. Horowitz draws a direct line from the Industrial Revolution to the present moment. America won the 20th century because it had superior technology; the AI revolution presents an identical fork in the road. He frames a16z's mission as answering one question — what can the firm do to help America win technologically — and argues that every decision, from portfolio construction to government engagement, flows from that north star. > *"And so when I think about our role in the industry, it's what can we do to help America win technologically?"* ## [04:04] American Dynamism, AI & Catching Up to China Ulevitch asks what has most surprised Horowitz about investing at the intersection of national security and venture capital since launching the American Dynamism practice. Horowitz explains why American-style freedoms are structurally irreplaceable: the Declaration of Independence's claim that rights are self-evident — not granted by government — makes them nearly impossible to revoke, a feature no other country has replicated at the same strength. On the competitive landscape with China, Horowitz notes the pre-ChatGPT conventional wisdom gave China a large AI lead, primarily because China had integrated AI deeply into its military and government bureaucracy while the U.S. lagged far behind. The most heartening development since then has been the speed of American catch-up: a wave of entrepreneurs willing to serve the national interest, combined with a U.S. government genuinely open to new companies and willing to change procurement rules to accommodate them. > *"But the the thing that was true about the kind of old incorrect idea was that they were way ahead of us in integrating um their AI technology with uh their government you know on a kind of military basis on a bureaucracy basis you know and all facets and so you know when we started we were coming from I would say very far behind you know in that you know in that idea um the thing that's been surprising though is like how fast um we've been catching up."* ## [08:50] The Anthropic Deal: What Really Happened The conversation turns to the high-profile collapse of Anthropic's contract with the Department of War. Horowitz offers a deal-mechanics reading that cuts through the public framing: Anthropic had overwhelming leverage — they were already deployed, the country was heading toward conflict, no software vendor has ever had more negotiating power — yet they walked away. In Horowitz's view, that behavior has only one explanation: Anthropic wanted out of the deal, likely due to internal employee pressure, and used a philosophical disagreement as the exit ramp. He pushes back on the framing that a national security AI contract is ethically compromised. The Department of War operates under more rules and oversight than any private entity, and leaks are effectively guaranteed if those rules are broken. Ulevitch extends the point to founders more broadly: companies that let employees veto geopolitical decisions are substituting "vibe geopolitics" for the considered judgment of people who have studied — and sacrificed for — these questions their entire careers. > *"It fell apart because Ananthropic wanted out of the deal."* ## [13:37] Exporting American Dynamism to Our Allies Ulevitch raises a geographic expansion question: American Dynamism's name is parochial, but the practice is really about America and its allies. Horowitz has spent significant time abroad meeting foreign leaders who want to replicate U.S. startup culture. He outlines why that's hard — entrepreneurship at scale requires a deep-seated belief that the government won't arbitrarily seize what you build, and very few countries (Sweden and Israel being notable exceptions) have that culture. He identifies concrete partnership opportunities: Mexico's high-quality manufacturing expertise in automotive and adjacent sectors; Japan's robotics heritage and surging defense spending (moving from 0% to 3% of GDP), which creates aligned interests given shared concern about China. The section closes with Ulevitch flagging that the coming robotics revolution will be the next major theme for the practice. > *"America does give everybody a chance and entrepreneurs can really count on that."* ## [16:56] Power, Responsibility & How a16z Serves Founders A recent profile described a16z as a "power broker" using capital and networks to shape markets. Horowitz reframes the description: power isn't something the firm accumulates for its own sake — it's a feature of the product offered to founders. Entrepreneurs have great ideas but lack the power to get the right meeting with Congress, secure a key enterprise customer, or navigate regulation; a16z's scale converts that gap into founder advantage. The internal culture is deliberately countervailing. The firm's first cultural principle — "first-class business, only in a first-class way" — means showing up on time, responding promptly, and being honest. These small behaviors prevent the firm from drifting into a posture where it treats founders as supplicants rather than partners. > *"So power is sort of a feature of our offering is the way I think about it."* ## [18:58] The State of Venture Capital & Why Most Firms Can't Scale Horowitz provides a structural explanation for why most venture firms cannot grow beyond a certain size. The original design premise of the industry was that only ~15 companies per year would ever reach $100 million in revenue, so small partnership structures with shared economics and shared control made sense. Mark Andreessen's "software is eating the world" thesis invalidated that premise: every company is now a technology company, so the target universe has exploded and so has the need for organizational scale. Scaling to capture that universe requires organizational reorganization — which requires a single decision-maker. Firms built on consensus control cannot reorg cleanly, because those who lose power in a reorg will block it. A16z, with centralized control from inception, was structured to reorg repeatedly and now fields 600+ people organized as small teams sharing a common platform. The result is a barbell: large generalist firms that cover every technology domain, and narrow specialists focused on AI infrastructure, bio, crypto, or games. The mid-size generalist firm is being squeezed out. > *"when you redistribute power, people are mad if they get a vote um that they're going to foul that that that reorganization and you can't scale without reorging."* ## [23:21] The New Rules of Media The media discussion opens with a structural observation: old and new media are not different games — they are the same game with different rules. Under scarcity (limited channels, rigid formats), the winning strategy was defense: avoid gaffes, because a Howard Dean scream lives forever on a three-channel media landscape. Under abundance (unlimited channels, unlimited formats), the winning strategy is offense: be interesting, because anything boring simply drowns in the noise. Horowitz points to Alex Karp as the exemplar of the new model: relentlessly entertaining, consistently on message (pro-America), and unafraid to be unpredictable. The flood-the-zone correction mechanism — do ten podcasts after a mistake — makes individual errors survivable in a way they never were in the old world. His coaching to founders: you cannot win by not losing anymore; you win by being worth paying attention to. > *"Um, and so the key to winning isn't not making a mistake, it's being interesting."* ## [26:22] America's AI Optimism Gap Horowitz names his biggest worry: a polling result showing that more than 70% of Chinese citizens are optimistic about AI while fewer than 30% of Americans share that sentiment. He attributes the gap to an American media culture that foregrounds AI risks — surveillance, job displacement, existential threats — while systematically underweighting the positive case. He contrasts this with Japan, where renewed enthusiasm for AI has reignited the entire startup ecosystem. His ask of founders, policymakers, and technologists in the audience: rebalance the narrative. AI will end traffic deaths, cure cancer, and eliminate poverty as we know it. These outcomes deserve as much airtime as the dangers. He closes with the analogy of fire — a technology capable of burning down a village that nonetheless heats homes and cooks food — arguing that managing dual-use risk is the normal condition of every transformative technology, not a disqualifying exception for AI. > *"We're going to cure cancer."* ## Entities - **Ben Horowitz** (Person): Co-founder and general partner at a16z; primary speaker throughout, drawing on experience as a founder, CEO, and venture capitalist. - **David Ulevitch** (Person): General partner at a16z leading the American Dynamism practice; hosts the conversation at the American Dynamism Summit in Washington, D.C. - **Andy Grove** (Person): Former CEO of Intel; Horowitz's mentor whose maxim on industry leadership frames the episode's opening section. - **Alex Karp** (Person): CEO of Palantir; cited as a model for direct, entertaining, on-message communication in the new media landscape. - **Mark Andreessen** (Person): Co-founder of a16z; author of "software is eating the world," the thesis underpinning a16z's scaling rationale. - **American Dynamism** (Concept): a16z's investment practice focused on companies serving U.S. national interests — defense, manufacturing, advanced software and hardware — now extended to allied nations. - **Anthropic** (Organization): AI safety company whose contract with the U.S. Department of War collapsed; Horowitz argues the deal fell apart because Anthropic chose to exit, not over genuine ethical conflicts. - **a16z** (Organization): Andreessen Horowitz; raised over $15 billion in its latest fund, the largest in firm history and the largest VC fund ever raised. - **Department of War** (Organization): U.S. federal defense department; counterparty in the Anthropic procurement deal and key customer for American Dynamism portfolio companies. - **Palantir** (Organization): Defense and analytics software company; referenced as an exemplar of a firm successfully working at the intersection of Silicon Valley and national security.

Dan Shipper interviews Angela Jiang (head of product) and Katelyn Lesse (head of engineering) for the Claude platform at Anthropic, recorded at the Code with Claude developer event. The conversation unpacks how Claude's platform has grown from a simple completion API into a fully managed agent infrastructure, why the harness and the model are increasingly inseparable, and what the "outcome + budget" vision means for the future of agent development. Together the three trace every stage of the agent lifecycle — from spinning up a first session to retiring stale agents — and share candid war stories from Anthropic's own internal deployments. ## [00:00] Where the platform will be in a year Dan opens with a question the rest of the episode keeps circling back to: a year from now, where is the platform? Angela's answer — Claude understands itself well enough to pick its own sub-agents and write its own harness on the fly. Katelyn picks up the other half: an infrastructure layer that can keep up with agents that continually rewrite themselves. This exchange actually comes from late in the interview; the show puts it up front because the whole conversation is about how today's primitives get you there. > *"We'd want to experiment with directions where Claude actually gets so good at understanding itself, it figures out what model you should be using, it figures out how to spin up all the sub agents."* — Angela Jiang ## [01:48] How the Claude platform evolved from API to agents Angela traces the arc from early LLM APIs — stateless, exploratory, maximum surface area — through session-based chat, and now into fully autonomous agents. The through-line is always the same: raise the abstraction layer high enough that customers can get the best outcome from Claude with as little work as possible. Early adopters wanted every raw knob; today, most teams arriving at Anthropic want a substantial set of things "out of the box." The platform's job is to keep shrinking the distance between intention and outcome. > *"It probably ends up just being like whatever it's like the set of primitives and infrastructure that enables you to basically get the outcome as fast as possible with actually as little of work as possible."* — Angela Jiang ## [04:09] The primitives that make up Claude Managed Agents Katelyn explains that Claude Managed Agents is assembled from the same primitives available to anyone on the Messages API — code execution sandboxes, web search, and built-in tools — but wrapped in a curated harness Anthropic has already battle-tested internally. Angela adds that the team is opinionated about two primitives in particular: file systems and skills. These are treated as load-bearing choices that shape how Claude behaves across all agent tasks. The platform is designed to be modular so developers can plug in custom pieces where the standard harness does not fit, and Anthropic publishes reference implementations for teams that want to stay on the Messages API directly. Dan describes his team running Claude via the `claude -p` command on Mac Minis and worries about lock-in and divergence from Claude Code. Katelyn responds that Anthropic's internal first-party products run on the same platform as external customers, which means divergence between Managed Agents and Claude Code will shrink over time. > *"We've taken what we see as all the most powerful of those things and put them together into a harness and a set of infrastructure that is just the way to get what we think is the best outcomes out of Claude."* — Katelyn Lesse ## [10:37] Why the harness and the model are becoming a single unit Angela challenges the conventional wisdom that a generic, model-swappable harness is the right architecture. As models diverge in technique across labs, the alpha is in tight harness-model co-design rather than hot-swapping. Internally, Anthropic tested multiple harness variants for the memory feature and found they performed "drastically differently." The implication: treat the agent (harness + model) as the unit of redundancy, not the model alone. Dan pushes on whether this creates path dependence in the model itself. Angela acknowledges that the primitives chosen really do shape the model's trajectory, and that being wrong about them is hard to undo. She cites models that over-indexed on reasoning versus those that went deep on computer-use as two diverging paths that are difficult to reverse. > *"The harness and the model get very paired. You still need redundancy, and you still might want to use other models for things, but you probably do it at the layer of like the agent, meaning like the harness plus the model."* — Angela Jiang ## [18:49] The infrastructure wall that kills most agent projects in production Katelyn identifies the real blocker for most agent projects: not harness engineering, but the infrastructure wall hit when teams try to move from prototype to production. Keeping a persistent server alive, managing sandbox failures, storing transcript data, and handling secure credential injection — these mundane concerns kill projects that technically "work" on a Mac Mini. Anthropic's own repeated experience of hitting this wall internally was the primary motivation for building Managed Agents. Angela describes the vaults primitive as an early step toward one-click agent deployment: once agent identity and credentials are handled securely at the platform layer, adding a Slack integration should eventually be as simple as telling Claude to "add Slack" and watching the bot appear. > *"Everyone hits the same problem of like, oh wow, I either need to like keep a server constantly running or I need to use infrastructure that will spin up and spin down, and I need to store the transcript data, and I need secure sandboxing, and all these sorts of things."* — Katelyn Lesse ## [24:49] Why team agents need a different shape than individual productivity tools Angela explains why individual productivity tools like Claude Code do not simply scale to team use. The moment three people want a shared agent that automates an end-to-end process across roles, a laptop-resident tool breaks down in availability, access control, and coordination. She cites Guillermo Rauch of Vercel's framing of an internal "AI software factory" as the right mental model: not individual augmentation, but a full organizational stack of agents that continuously produces high-leverage output for every function in the company. > *"When you get to the team layer suddenly everything gets like massively more complex. Like number one obviously it can't like sit on your laptop."* — Angela Jiang ## [26:36] How Anthropic's legal team uses an agent to review marketing copy Katelyn walks through one of Anthropic's own internal deployments: a legal-review agent that accepts marketing copy submissions and performs a first-pass review before anything reaches a human lawyer. The agent can approve copy outright or escalate for human review, eliminating low-value ticket-queue work. The form factor is a thin app layer on top of Managed Agents with shared visibility across both teams. Angela and Dan dig into why this is an agent rather than a skill: human-in-the-loop requirements, the need to spin up separate sessions, and multi-team collaboration all exceed what a single skill invocation can handle. The governance model that emerged was notable: rather than gating changes behind the platform team, end users discovered they could self-serve small improvements via Claude Code. Angela describes the end-state user experience as simply "talking to Claude," even when the underlying system is "many many Claudes engaging with each other." > *"Under the hood it's many many Claudes engaging with each other to get to the part where then they the Claudes themselves are doing the more complex work that the human doesn't really necessarily need to interpret."* — Angela Jiang ## [34:24] Using multi-agent orchestration for advisor strategies, adversarial pairs, and swarms Angela highlights three multi-agent architecture patterns people are assembling with the newly launched orchestration primitives: an advisor strategy that separates execution from advice; adversarial pairs where one agent generates and another critiques; and swarms that split a problem into many small parallel pieces and recombine results. Each pattern suits a different problem class — swarms excel at bug hunting, while wide-research tasks benefit from advisor or parallel-decomposition architectures. LEGO-like primitives let practitioners hill-climb at the architecture level, not just the prompt level. > *"If we can make the primitives very LEGO-like, then people can put them together to solve things at a slightly higher form factor, which is more like an architecture or like a strategy."* — Angela Jiang ## [35:50] How to measure agent success with outcome and budget as the end state Angela frames the long-term measurement philosophy: compress everything to an outcome and a budget, and let the platform resolve all intermediate decisions. Domain-specific evals (e.g., PR-merge rate for coding agents) remain useful today, but the target is a verifiable outcome spec that Claude can grade itself against repeatedly. Katelyn addresses the adjacent problem of agent staleness: Anthropic has built skills to help teams upgrade agents when new models ship, and the most forward-leaning teams already run meta-agents that monitor other agents for degradation and trigger upgrades automatically. > *"Our kind of principle of like maybe the end state of some of these things is that everything should kind of compress down to an outcome and like a budget. And that's probably like about it."* — Angela Jiang ## [39:11] What the platform looks like a year from now, when Claude writes its own harness Angela envisions a world where users supply only an outcome and a budget, and Claude self-selects models, spins up sub-agents, and writes its own harness on the fly — eliminating harness engineering entirely, just as today's platform has already eliminated much of manual tool construction and prompt engineering. She is cautiously optimistic that the "outcome" half of the equation may be achievable within a year with some budget error bars. Katelyn adds the infrastructure corollary: such a world requires a platform capable of supporting agents that continuously recreate themselves, handling arbitrarily shaped long-running requests without ever becoming the bottleneck. > *"Claude is actually able to understand itself enough that it can come almost like write itself on the fly to figure out what is necessary in that kind of like two-parameter world of like outcome and budget."* — Angela Jiang ## Entities - **Angela Jiang** (Person): Head of Product for the Claude platform at Anthropic; co-architect of the Managed Agents product vision. - **Katelyn Lesse** (Person): Head of Engineering for the Claude platform at Anthropic; focuses on infrastructure reliability and scale. - **Dan Shipper** (Person): Host of AI & I on Every; CEO of Every; building internal agent products on the Claude platform. - **Claude Managed Agents** (Software): Anthropic's hosted agent infrastructure — a harness plus cloud compute that wraps the Messages API with built-in memory, sandboxing, vaults, and skills. - **Messages API** (Software): Anthropic's core API; the underlying primitive on which Managed Agents and all first-party products are built. - **Anthropic** (Organization): AI safety company that builds and operates the Claude model family and its associated platform. - **Every** (Organization): Media company producing AI & I; an early Managed Agents customer building internal editorial agents. - **Stripe Minions** (Software): Stripe's internal end-to-end software development platform built on agent infrastructure; cited as a model for company-wide coding agent deployment. - **Vercel** (Organization): Developer infrastructure company; CEO Guillermo Rauch's "AI software factory" framing used as the mental model for team-level agent adoption. - **Outcome + Budget** (Concept): Anthropic's long-term design principle that the final form of agent interaction should require only a verifiable outcome and a cost ceiling, with the platform resolving all intermediate decisions.

In one of their most consequential episodes, the All-In besties dissect SpaceX's surprise compute lease to Anthropic — the deal that may cement Anthropic as AI's dominant platform — and debate whether David Sacks's "Rockefeller" framing is prophecy or paranoia. The group then wrestles with a White House trial balloon about an "FDA for AI," ultimately concluding it was mostly media spin, before closing with a bullish-but-cautious read on the AI-driven market boom. Brad Gerstner fills in for David Friedberg, bringing investor perspective from both public and private markets across the episode's 82 minutes. ## [00:00] Bestie intros! Thoughts on the LA mayor election Jason Calacanis opens with the full crew: Chamath Palihapitiya, David Sacks, and fifth bestie Brad Gerstner joining in for David Friedberg, who is out sick. The warm-up quickly turns to the LA mayoral race, where Spencer Pratt is mounting a surprisingly effective challenge to incumbent Karen Bass. The group praises Pratt's viral debate performance — evisceration of the city council candidate over homeless policy — and Chamath notes the power of a sharp social-media team in modern politics. Brad flags a California ballot initiative that would constitutionally protect retirement savings and ban a wealth tax, reading it as a potential seismic signal. Jason observes that New York City hedge-fund titan Ken Griffin publicly announced he is pulling investment from New York after NYC councilman Zohran Mamdani targeted his home in a campaign video, underlining the tension between aggressive progressive politics and capital flight. > *"If California effectively passes a constitutional amendment protecting retirement savings and personal assets and banning the wealth tax and [Spencer Pratt] gets elected, the message that would send to the country — that's a very non-consensus view that I'm becoming increasingly optimistic about."* — Brad Gerstner ## [04:38] SpaceX-Anthropic deal, Elon Web Services, SpaceX IPO valuation, Anthropic's insane growth trajectory Jason leads with the blockbuster news: SpaceX has leased all of Colossus 1 — its H100-based Memphis data center — to Anthropic, adding over 220,000 Nvidia GPUs and 300 megawatts to Anthropic's supply-constrained capacity. The deal immediately doubled Claude Code's rate limits and removed peak-usage caps for paid users. Chamath frames Anthropic's explosive growth as purely supply-constrained: if unlimited power existed, revenues would be "even more parabolic." He sees the deal as Elon strategically de-risking SpaceX's valuation story — blunting bear cases around delayed orbital data centers while generating near-term revenue to subsidize Grok training. Brad estimates the arrangement adds $4–5 billion in incremental 2026 revenue for SpaceX, calling EWS (Elon Web Services) a genuine fourth hyperscaler alongside AWS, Azure, and GCP. He also warns that organized activists — not organic local opposition — are using the same playbook that stalled nuclear construction in America to delay data-center permitting. David Sacks notes that Anthropic grew from $10B ARR on January 1 to $44B ARR by April — a trajectory he calls unlike anything Silicon Valley has ever witnessed. > *"Nobody in Silicon Valley has ever seen anything like it. Forget about the rest of the country. I mean, all we do in Silicon Valley is deal with exponentials. And still, people have never seen that kind of growth at that level of scale."* — David Sacks ## [26:48] Is Anthropic the next great monopoly? Early signals or major overreaction? David Sacks draws an extended analogy between Anthropic and John D. Rockefeller's Standard Oil, arguing that safety-first rhetoric can function as regulatory capture — building a moat that locks in the emerging duopoly of Anthropic and OpenAI while blocking competitors. He notes that if Anthropic sustains its 10× annual growth for just 18 more months it could become "the most powerful monopoly ever created in human history," dwarfing the combined Mag-7 revenue. Brad pushes back hard: Anthropic and OpenAI are still fledgling startups on a GAAP basis, Google and Amazon are producing hundreds of billions in free cash flow to fund competing models, and pre-emptive antitrust action at the starting line of AI would be "a disaster." Jason translates Brad's position as "don't mess with my paper," since Altimeter holds positions in several of these companies. Sacks clarifies his northstar is vigorous competition — but he flags Anthropic's banning of OpenClaw from using its API as a concrete anti-competitive act worth scrutiny. > *"Unless something about their current trajectory changes, Anthropic will be the most powerful monopoly ever created in human history — a trillion dollars of ARR growing at some rate. Dario calls it AGI. I call it the biggest monopoly in human history."* — David Sacks ## [35:21] "FDA for AI" freakout, how the White House thinks about AI safety Reports surfaced that the White House was considering an executive order to create an AI working group that could require pre-release safety reviews for new frontier models — triggered, according to the New York Times, by Anthropic's classified "Mythos" model reportedly alarming national-security officials. NEC Director Kevin Hassett appeared on Fox Business drawing an FDA analogy, while Treasury Secretary Scott Bessent spoke more carefully about balancing innovation and safety. Sacks calls much of it "fake news" amplified by Andrew Ross Sorkin's DealBook column, noting that Susie Wiles, the White House Chief of Staff, issued a statement walking back the FDA framing. He reveals he spoke with Hassett directly and confirms no senior official actually supports a pre-approval regime. He points to the White House's March 20 National AI Regulatory Framework as evidence the administration favors specific solutions over broad regulatory capture. The group converges on one concrete measure: KYC (Know Your Customer) requirements before frontier model API access during preview periods, plus rapid deployment of cyber-capable AI to companies like CrowdStrike and Palo Alto Networks. > *"There is a substantial faction of AI ideologues or doomers who are basically employing the classic 'never let a crisis go to waste' strategy. Yes, we do have this cyber issue that is real — everyone needs to harden their systems now. But what they're trying to do is use that issue to try and create a permanent new infrastructure in Washington."* — David Sacks ## [52:01] Flipping AI's negative perception: Giving, healthcare and education innovation Jason shifts from regulatory defense to offense: how should the tech industry proactively counter negative public perception of AI? He proposes that companies going public — Anthropic, OpenAI, SpaceX — could dedicate 1–5% of IPO proceeds to every American via "Invest America" accounts, creating tangible shared upside. He also calls for serious engagement on minimum wage and universal healthcare, arguing that a financially healthier consumer base is structurally good for capitalism itself. Brad endorses the "Invest America" concept, adding that data center host communities should receive direct benefits like free local electricity. David pivots to political salience data: AI ranks 29th out of 39 voter issues — well below cost of living and economic growth, two metrics where AI is actively deflationary and expansionary. The industry's real message should be economic delivery, not safety governance. Chamath gives tech leaders a "D-minus trending to F" for communications and calls for tangible reinvestment in America at scale. > *"I think that there's a pretty profound vibe shift with respect to tech, tech oligarchs, Silicon Valley, and particularly AI. That vibe shift has already happened on Main Street, and I think that's starting to seep into Washington."* — Chamath Palihapitiya ## [60:04] Trading the AI market, state of the economy Brad leads a comprehensive market check: AWS on a $150B run rate (28% growth), Azure at $108B (39%), Google Cloud at $80B (63%). The S&P 500 is at all-time highs, the 10-year sits at 4.3%, and inflation is under control — far better outcomes than the doom scenarios predicted around tariffs and geopolitical conflicts. S&P 500 operating margins improved from 11% in 2023 to 13% in Q1 2026, and the Mag-5's combined headcount grew only 3% over three years while revenues surged. Chamath urges caution: there is still no direct evidence AI is lifting enterprise profit margins in aggregate, and a reckoning arrives in roughly 500 days when the fork between opex reduction and revenue growth will determine whether the AI boom is real or a mirage. Jason counters that for startups the ROI is already "fait accompli" — AI-generated ad creative at Nike and DoorDash, portfolio companies shipping product at half the headcount. David credits Trump administration policies — rescinding Biden's chip-export licensing and AI-approval regime, unleashing energy permits — for creating the conditions that enabled the boom, and notes that the unemployment rate for recent college graduates has actually improved, contradicting the entry-level-job-loss narrative. > *"I think we have kind of call it 500 days where you just got to be net long. But I think it's literally in the hundreds of days from now that you're going to have to have an important reckoning moment. The people that are paying for all these tokens need to see an actual benefit."* — Chamath Palihapitiya ## Entities - **Jason Calacanis** (Person): Host and moderator; angel investor and podcast co-founder - **Chamath Palihapitiya** (Person): General partner, Social Capital; co-host; contrarian macro voice on AI ROI and market cycles - **David Sacks** (Person): Co-host; former White House AI & Crypto Czar; framed Anthropic as a potential historic monopoly using the Rockefeller analogy - **Brad Gerstner** (Person): Founder & CEO, Altimeter Capital; fifth bestie; bullish on compute stocks and AI market structure - **Dario Amodei** (Person): CEO of Anthropic; referenced as "Daario D. Rockefeller" by Sacks; party to the SpaceX compute deal - **Elon Musk** (Person): CEO of SpaceX and xAI; architect of Elon Web Services and the Colossus 1 compute lease strategy - **Anthropic** (Organization): AI lab behind Claude; grew from $10B to $44B ARR in four months; center of monopoly and FDA debates - **SpaceX / xAI** (Organization): Lessor of Colossus 1 data center to Anthropic; emerging fourth hyperscaler under EWS branding - **Elon Web Services (EWS)** (Concept): SpaceX's compute-leasing business positioned as a hyperscaler competitor to AWS, Azure, and GCP - **Mythos** (Software): Anthropic's classified cyber-capable frontier model that reportedly alarmed White House national-security officials - **KYC for AI** (Concept): Proposal to require identity verification before granting API access to frontier models during preview periods - **Invest America** (Concept): Proposal for IPO-stage tech companies to dedicate a share of proceeds to universal investment accounts for US citizens

Un court tutoriel Anthropic sur les hooks Claude Code, la porte de sortie déterministe pour tout ce qui doit absolument s'exécuter à chaque édition, chaque appel d'outil, chaque commit. Le message clé : si vous écrivez "toujours exécuter prettier" dans claude.md en espérant que ça marche, vous avez déjà perdu. Mettez-le dans un hook. ## [00:02] Ce que sont les hooks et pourquoi ils sont déterministes Les hooks se déclenchent à des points fixes dans le cycle de vie de Claude Code, et l'argument central du narrateur est que, contrairement aux instructions au niveau du prompt, ils s'exécutent toujours. Indiquer au modèle dans claude.md de lancer prettier après chaque édition fonctionne la plupart du temps, mais "la plupart du temps" est exactement la faille que comble un hook. Même intention, mais imposée par le runtime plutôt que suggérée au LLM. > *You can tell Claude in your claude.md file to run prettier after every file edit and most of the time it will do that, but sometimes it won't. It's not perfect. But a hook makes it happen every single time with no exceptions.* ## [00:37] Cas d'usage courants Quatre exemples représentatifs délimitent le périmètre : formatage automatique après les éditions de fichiers, journalisation de toutes les commandes exécutées pour la conformité, blocage des opérations dangereuses comme la modification des fichiers de production, et envoi d'une notification lorsque Claude termine une longue tâche. > *Common use cases could include auto formatting after file edits, logging all executed commands for compliance, blocking dangerous operations like modifying production files, and sending yourself notifications when Claude finishes a task.* ## [00:52] Configurer les hooks et les cinq événements du cycle de vie La configuration réside dans `settings.json` : choisissez un événement, affinez-le optionnellement avec un matcher pour l'outil concerné, puis fournissez une commande shell. Cinq événements couvrent la boucle : `UserPromptSubmit` avant que Claude voie un prompt, `PreToolUse` et `PostToolUse` encadrant chaque appel d'outil, `Notification` quand Claude alerte l'utilisateur, et `Stop` quand Claude termine sa réponse. > *Pre-tool use which runs before a tool call, post-tool use runs after a tool call completes. Notification runs when Claude sends a notification, and stop runs when Claude finishes responding.* ## [01:22] Formatage automatique avec un hook post-tool-use L'exemple canonique : un hook `PostToolUse` avec un matcher `Edit` ou `MultiEdit` se déclenche chaque fois que Claude modifie un fichier. La commande vérifie l'extension et dirige vers le bon formateur : prettier pour TypeScript, gofmt pour Go, ruff pour Python, ou tout ce que le projet standardise. > *You set a post-tool use hook with a matcher of edit or multi-edit, right? So, it fires whenever Claude modifies a file. The command checks the file extension and runs the appropriate formatter.* ## [01:49] Bloquer les appels d'outils avec pre-tool-use et les codes de sortie Les hooks `PreToolUse` reçoivent le nom de l'outil et son entrée en JSON sur stdin, puis décident via le code de sortie : `0` pour continuer, `2` pour bloquer. Quand un hook bloque, ce qu'il a écrit sur stderr est renvoyé à Claude comme feedback, permettant au modèle de comprendre pourquoi et d'adapter son plan. C'est ici qu'on applique les règles strictes : bloquer les écritures dans un répertoire de config de production, refuser les commandes bash contenant `rm -rf`, bloquer les commits sur main. Le narrateur résume : ce que l'équipe doit garantir, pas simplement suggérer. > *If it exits with code two, the action is blocked and the STD error message gets fed back to Claude's feedback so Claude knows why it was blocked and can adjust.* ## [02:26] Hooks au niveau projet et partage en équipe Les hooks dans `.claude/settings.json` ont une portée projet et peuvent être commitués dans le dépôt, ce qui signifie que toute l'équipe les hérite automatiquement au clonage. Référencez les scripts via la variable d'environnement `CLAUDE_PROJECT_DIR` pour que les commandes se résolvent correctement, quel que soit le répertoire courant de Claude. La règle de clôture : si quelque chose doit se produire à chaque fois sans exception, ne le mettez pas dans un prompt, mettez-le dans un hook. > *If something needs to happen every time without fail, don't put it in a prompt. Put it in a hook.* ## Entities - **Anthropic Tutorial Narrator** (Person): La voix officielle d'Anthropic pour la série de tutoriels Claude Code 101. - **Claude Code** (Software): L'outil de codage terminal agentique d'Anthropic auquel les hooks se connectent lors des événements du cycle de vie. - **Hooks** (Concept): Commandes déterministes se déclenchant à des points fixes dans la boucle Claude Code, alternative imposée par le runtime aux instructions au niveau du prompt. - **settings.json** (Configuration): Là où les hooks sont déclarés ; `.claude/settings.json` à la racine du projet est versionné dans le dépôt afin que les équipes partagent les mêmes règles. - **PreToolUse / PostToolUse / UserPromptSubmit / Notification / Stop** (Events): Les cinq événements du cycle de vie auxquels un hook peut se rattacher. - **CLAUDE_PROJECT_DIR** (Environment variable): Utilisée dans les commandes de hook pour référencer des scripts relatifs au projet, indépendamment du répertoire courant de Claude.

Matt Pocock joins swyx at AI Engineer Europe to argue that the old software design canon — DDD, deep modules, ubiquitous language — matters more, not less, in the AI coding era. The thesis: code is not just a compile target; a codebase that is easy for humans to change is easy for AI to change. Along the way they cover course-making, why traditional lectures still beat AI-native learning, and TypeScript's quiet takeover of AI engineering. ## [00:04] Opening at AIE Europe and the Cursed Course swyx welcomes Matt to the AI Engineer Europe podcast booth in London. Matt jokes that AIE is "the worst" event he has ever attended (the location is in fact astonishing) before turning to his Claude Code course, which is just wrapping up its two-week cohort. He explains why he runs short cohorts: AI moves so fast that self-paced courses cannot guarantee updates, and the "curse" of releasing into breaking changes — AI SDK v5 dropped on day two of his AI SDK v4 course, and the Claude Code source leaked during this one — is now baked in. The conversation then turns to teaching as a craft. Matt rejects the "pundit" branch of YouTuber identity — he is not trying to predict the future, only to teach durable material — and notes that being a teacher first is what differentiates his content. > *I'm not a guy who's trying to predict the future. I'm just trying to teach.* ## [02:51] Why Engineering Fundamentals Matter More with AI Matt previews his AIE talk. The popular narrative says code no longer matters because English plus an AI compiler can produce applications. Every time he tried to ignore the code, he ended up with "a terrible mess." So he went back to the classics — *Extreme Programming*, *The Pragmatic Programmer*, *A Philosophy of Software Design*, DDD — and discovered they ported directly into prompts. Keeping the architecture in your head, even when you delegate implementation, yields outsized dividends. > *If you have a code base that's easy to change for humans, it's going to be easy for AI to change, too.* ## [04:23] Narrow Waist and Deep Modules swyx introduces the "narrow waist" concept from internet architecture (TCP/IP, HTTP at layers 3–4) as a way to contain AI-generated slop: define rigid interfaces, delegate the inside. He extends it to running AIE as a nine-person business — "model-view-claw" instead of MVC, where coordination across people and AI is the real systems problem. Matt maps this onto John Ousterhout's notion of *deep modules*: a large amount of functionality behind a simple interface, ports and adapters style. This is, in his experience, the best way to use AI for coding — be intentional about the interface as a human, then delegate the implementation. > *Deep modules basically — a large amount of functionality with a simple interface. Kind of ports and adapters, right?* ## [06:37] Domain-Driven Design Meets AI DDD is having a moment, and Matt argues it works *because* the framework has been around long enough to sit in the latent space of these models. You do not have to invent new vocabulary; you can bolt on a system that is composable and that the model already understands. The deeper point: DDD is fundamentally about aligning code with language, which is exactly what you want when speaking to an AI. He makes it concrete with the `mattpocock/skills` repo (≈13k stars) and its "ubiquitous language" skill — a Claude Code skill that scans your codebase, surfaces the arcane jargon, and refines it with you into a markdown file he keeps open while prompting. He references it from `agents.md` but does not paste it wholesale, so the agent finds it when searching for those terms. > *Essentially, you're trying to create a unified domain model so that the AI and you are speaking the same language.* ## [10:05] Teaching as an Overpowered Skill swyx asks how Matt got so good at explaining things. Matt credits six years as a voice coach before becoming a developer — communication felt like an unfair advantage when he started as a junior. He has since narrowed his focus: split time between learning material and finding the right phrases for it. The old texts help because they give him pre-built mental models to explain new ideas through. He walks through his course-making process: an "explore and exploit" phase, a Zettelkasten-style Obsidian vault, a custom planning app, P1/P2/P3 prioritization, and the rule that *each lesson teaches exactly one thing* with dependencies made explicit. Most of what he produces ends up on the cutting room floor. > *The ability to communicate always just felt like a ridiculous overpowered skill that I had in my locker that no one else had.* ## [13:20] How People Actually Learn AI Engineering The conversation turns to whether AI has changed how people learn. Matt distinguishes knowledge (lectures), skills (interactive exercises), and wisdom (small-group discussion — and now, talking to an AI). Counterintuitively, the more he leans into AI-experimental teaching, the more it turns his audience off. Most learners still want traditional lectures; swyx recalls Maven's cohort-based education arc landing in the same place. Matt's compromise is to force the work without forcing the form: in his TypeScript material he throws learners into a problem first and gives them the knowledge afterwards. > *The more I lean into the kind of AI experimental stuff, the more it actually turns people off my materials.* ## [15:04] TypeScript Overtaking Python swyx flags that TypeScript overtook Python in the GitHub survey this year — a shift he did not see coming, particularly in AI engineering where Python's expressiveness has been dominant on the backend. Matt's echo chamber is 100% TypeScript, but his real argument is ecosystem: when you care about UX and shipping chat-style applications, the framework gravity is in TypeScript (Vercel's Next.js, Cloudflare's variants). swyx admits this would meaningfully change which frameworks he promotes. > *If you're concerned about UX, concerned about shipping great stuff, you're mostly doing it in TypeScript.* ## [16:45] Inversion of Control and Composable Skills Matt looks ahead. His TypeScript-evals bet (Everlight) stalled — "no one's excited to do evals." The next frontier is *inversion of control*: as coding agents converge on similar architectures (Firebase-style backends, small tool sets), the interesting axis becomes how much control sits with the developer versus the harness. Claude Code's opacity buys ease of use but loses observability; Pydantic AI ("Pi") swings the other way — total control, total maintenance burden. He closes by pointing past coding agents entirely. Software engineers are a step ahead because AI produces quality output in their domain, but the composable skills he authors — like his three-sentence "grill me" skill that makes the AI interrogate you until you reach a shared understanding — generalize to any domain where you want the AI aligned with you. > *The inversion of control is going to be really important — you put more control in the hands of the developer and less in the harness.* ## Entities - **Matt Pocock** (Person): Creator of Total TypeScript and AI Hero; teaches TypeScript and AI Engineering through two-week cohort courses. - **Shawn Wang / swyx** (Person): Host; founder of AI Engineer and the AIE conference series. - **AI Engineer Europe (AIE)** (Organization): The London conference where this conversation was recorded; Matt's talk hit 1M views in 13 days — fastest in AIE history. - **AI Hero** (Organization): Matt's AI engineering education platform (aihero.dev). - **Claude Code** (Software): Anthropic's coding agent; subject of Matt's just-finished course and a recurring example throughout. - **Domain-Driven Design (DDD)** (Concept): Software methodology centered on aligning code with the language of the business domain; Matt argues it ports cleanly into AI prompting. - **Ubiquitous Language** (Concept): DDD practice of maintaining a shared vocabulary doc; Matt's namesake Claude Code skill scans a repo and refines this with the user. - **Deep Modules / Narrow Waist** (Concept): Architectural pattern (Ousterhout / internet protocols) of large functionality behind a small interface — Matt's preferred shape for AI-assisted codebases. - **mattpocock/skills** (Software): Matt's open-source repository of Claude Code skills; ≈13k stars at recording time. - **Pydantic AI (Pi)** (Software): Python agent framework built from low-level primitives; cited as the high-control counterpoint to Claude Code's opaque harness. - **Obsidian** (Software): Note-taking app reportedly run by a team of four; the example for non-engineering domains where AI leverage compounds.

Dan Shipper and Austin Tedesco, Every's head of growth, discuss why the Codex desktop app has become their primary interface for all knowledge work — from drafting go-to-market plans to building live KPI dashboards — displacing Claude Code after months of side-by-side use. Dan frames the shift as the emergence of a new "agent management interface" operating system, while Austin walks through his live Codex setup in a screen-share session that covers automations, specialized agent suites, and recruiting workflows. The episode doubles as a practical field guide for non-engineers who want to run the same playbook. ## [00:00] A new operating system for knowledge work Dan opens cold: three months ago Codex was trash. Now Austin is the one firing it up before anything else each morning and routing 80 percent of his working time through it. Dan reads what changed structurally: a general-purpose coding agent that can reach into your filesystem, browser, and connected apps is becoming the operating system for knowledge work, and every major lab is racing for that surface. > *"There's a new operating system for how and where you're going to get your work done and it's this kind of agent management interface."* — Dan Shipper ## [00:57] How Codex went from a tool for senior engineers to a daily driver for knowledge work Dan traces the arc of Codex from its original positioning as a sandboxed pair-programming tool for senior engineers — one that "would argue with you, it would make you feel stupid" — to today's desktop app built on GPT-5.5. He attributes the pivot to OpenAI watching Anthropic prove with Claude Code that an emotionally intelligent, fast, computer-native agent creates a step-change experience for programmers and knowledge workers alike. The race is now between model companies to own the agent management desktop: Anthropic has Claude Code and Claude.ai desktop, OpenAI has Codex, and xAI has effectively acquired Cursor. ## [02:42] How Claude Code proved that a great coding agent works for any knowledge work Dan explains the insight that changed everything: if an agent can write software autonomously, it can do any kind of knowledge work autonomously. Claude Code demonstrated this first, drawing non-engineers — including Austin — into an agent-first workflow. OpenAI's hard pivot on Codex over the last three months is a direct response to that proof point. Dan describes the new paradigm as one where your agent is your interface to software, the internet, and daily tasks, not just a code co-pilot. > *"If it can write software on its own, it can do any kind of knowledge work on its own."* — Dan Shipper ## [07:24] Austin's switch to Codex Austin recounts his agent-pill moment: spending a December week inside Claude Code CLI, hooking it up to every tool he uses for work and personal life, and finding it indispensable for strategic thinking, data analysis, and drafting marketing copy. His initial Codex trial two months later felt alienating — the model was condescending, asking "Why?" when he requested clearer explanations. He kept Claude Code for 80 percent of knowledge work while tolerating Codex for engineering. The turning point was getting early access to GPT-5.5: at model parity, the decisive edge was the Codex desktop app itself — faster, better-organized, and with sub-agents that "just work." > *"So the idea that the codeex app is maybe 30 to 40% better is like that's a lot of work."* — Austin Tedesco ## [13:48] How Austin set up Codex with folders, keys, and reviewer agents Austin shares his screen and walks through his "Every Growth OS" folder inside the Codex app: a directory containing API keys for every tool the company uses (Gmail, Slack, Notion, Stripe), a CLAUDE.md project context file synced to GitHub, and a set of custom reviewer agents forked from Kieran Classen's Compound Engineering plugin. Where the standard Compound Engineering reviewers focus on security and front-end design, Austin's fork — publicly available as "Compound Knowledge" — reviews for strategic alignment with company goals and data accuracy, making it fit for knowledge-work plans rather than code PRs. The folder architecture lets Austin move seamlessly from a go-to-market draft to shipping a code PR without switching apps. > *"It's connected to everything we use for every and then some project instructional files that explain what the every business is, what we care about, how we like to work together."* — Austin Tedesco ## [18:24] Using Codex to brainstorm automations across Gmail, Slack, and Notion Austin demos his recommended on-ramp for new Codex users: open a fresh chat inside the Growth OS folder, run the Compound Engineering brainstorm workflow, and prompt the model to look at Gmail, Slack, and Notion and suggest automations. Codex surfaces a "follow-up radar" that triages incoming communications across sources, a command-center view for events and camps, and a recruiting pipeline automation — all calibrated to Austin's actual work context. Within the session, Codex writes automation scripts that require almost no tweaking and begins scheduling them; Austin highlights a nightly draft-reply routine that compiles unanswered messages and prepares replies for a quick thumbs-up approval. > *"They require very little tweaking to be like this is a thing I would and do use every day of there's this set of instructions that it comes up with based on what it knows about me."* — Austin Tedesco ## [22:42] How Austin manages the human review step when Codex is drafting communications A live audience question from Margaret prompts Austin to describe his human-in-the-loop review discipline. All drafting and orchestration happens inside Codex, but the final review intentionally lives in the native app: Slack draft replies are reviewed in Slack's drafts tab; email drafts are reviewed in Gmail; strategic plans are reviewed in Notion or the Proof markdown viewer. Stepping out of the agentic interface "freshens up my brain" before anything goes to a human. A second question from musician Alex about protecting high-value client emails leads to a discussion of how Austin uses Every's Kora email assistant together with Codex-managed rules, including having the agent interview the user to derive email rules rather than asking the user to specify them manually. > *"I just like for like the last pass before humans engage with it to step away from this agentic space and have a final check in another surface."* — Austin Tedesco ## [28:54] Using Codex to build specialized agents inspired by product executive Claire Vo Austin describes being inspired by a Claire Vo interview with Lenny Rachitsky in which Vo credited a suite of six specialized OpenClaw agents — rather than one overloaded master agent — as the key to unlocking leverage. Austin pasted the transcript of that interview directly into Codex and prompted it to propose six agents tuned to the Every growth function, provisioned into the company Slack. The agents occasionally break, but debugging is straightforward: screenshot the broken output or @-mention the Slack thread inside Codex and ask it to fix the agent's architecture. The result is a self-correcting loop where agent failures become Codex tasks. > *"Um I I actually just sent it the transcript of Claire's interview with Lenny and said like I want to do this too given everything you know about me and my work."* — Austin Tedesco ## [31:09] Synthesizing meeting transcripts and Slack threads into a go-to-market plan Austin walks through his most time-saving workflow: assembling a go-to-market plan for Every's upcoming Plus One product launch using nothing but Codex running the Compound Engineering brainstorm step against all existing meeting transcripts stored in Notion and Slack threads. With only five-minute windows between meetings, Austin prompted Codex to check the scheduled content calendar (a step it skips unless reminded), generate a proof doc, and push the final plan to Notion. The result was 80–90 percent complete. Dan adds the normative point: he prefers reading AI-written documents because they're easier for colleagues to produce, and the standard at Every is that you stand fully behind whatever your agent writes. > *"It's that I'm relying on the model to um look at all of the things that we've already said and thought about the go to market strategy, piece it together, and then review it, right?"* — Austin Tedesco ## [40:15] Building a live KPI tracker in Notion that agents can read Austin shares a more technical workflow: rebuilding Every's KPI tracker as a Notion database that updates every six hours by pulling from Stripe, social platforms, and other data sources via Notion's Workers tool. The tracker is explicitly designed to be both human-readable and agent-readable, so any team member's agent can query it and take autonomous actions — such as spinning up landing pages if an SEO keyword is underperforming. The challenge: the model can't one-shot the full tracker because even a 3–5 percent error in the MRR number is unacceptable for business decisions, so Austin is validating it column by column. Dan notes the philosophical complexity of defining revenue metrics consistently. > *"And so I have been doing this big kind of like to me complex uh workflow problem in codeex of let's build this sheet together, let's have it live in a notion database that all of our agents can point at."* — Austin Tedesco ## [44:54] Using Codex for recruiting Dan describes using Codex for outbound recruiting: he asked Codex to compile a list of General Assembly alumni and then filter it for people who had subsequently moved into AI, targeting candidates for an L&D director role. The first name on the resulting list was someone Dan considered a perfect fit who already followed him on Twitter, allowing an immediate DM. The section expands into a broader Q&A: Austin discusses when to fork Compound Engineering versus using it out of the box, how the team uses a shared Notion "compound" database to capture session learnings and turn them into reusable skills, and how Every's "Think Week" — a bi-annual week with no day-to-day work — creates organizational space for deep AI exploration. > *"Especially for any kind of like outbound effort, it can kind of find that needle in the haststack that you're looking for really really well."* — Dan Shipper ## Entities - **Dan Shipper** (Person): Co-founder and CEO of Every; host of the AI & I podcast; author of essays on AI and vibe coding - **Austin Tedesco** (Person): Head of growth at Every; Codex power user who manages the Growth OS project and suite of specialized agents - **Claire Vo** (Person): Product executive whose interview about specialized agent suites inspired Austin's multi-agent setup at Every - **Kieran Classen** (Person): Engineer at Every; creator of the Compound Engineering plugin used as the basis for Austin's knowledge-work fork - **Codex** (Software): OpenAI's desktop agent app, the primary tool discussed; runs on GPT-5.5 and supports sub-agents, folder-scoped projects, and plugin integrations - **Claude Code** (Software): Anthropic's CLI-based coding agent; Austin's previous daily driver before switching to Codex - **Compound Engineering** (Software): Plugin workflow framework by Kieran Classen; provides structured brainstorm, plan, and review steps used across Claude Code and Codex - **Every** (Organization): AI-focused media and software company publishing essays, courses, and tools; runs the AI & I podcast - **OpenAI** (Organization): Creator of Codex and GPT-5.5; provider of the ChatGPT Pro subscription whose credits were offered to camp attendees - **Notion** (Software): Primary knowledge-management and document platform at Every; used for meeting transcripts, the KPI tracker, and agent-readable databases - **GPT-5.5** (Software): OpenAI model powering the current Codex desktop app; reached parity with Claude Opus for Austin's knowledge-work tasks

Lex Fridman sits down with Jean-Baptiste Kempf, president of VideoLAN and lead developer of VLC, and Kieran Kunhya, longtime FFmpeg contributor and the voice behind the infamous FFmpeg account on X, for a four-hour deep dive into the invisible machinery behind virtually all video on the internet. Together they trace the full arc from raw bytes and container formats through hand-written assembly and codec reverse-engineering, confronting the open-source sustainability crisis along the way. The conversation is both a technical masterclass and a meditation on why brilliant volunteers—many of them teenagers—quietly build infrastructure that powers billions of devices every day. ## [00:00] Episode highlight The episode opens with a rapid-fire highlight reel that captures the spirit of what follows. Kempf distills the FFmpeg community's core value: code quality is the only credential that matters—"Maybe you're a dog. I don't care. I need to look at your code." Kunhya adds the scale: FFmpeg is running on roughly 100 million CPUs at any moment, with three billion devices continuously decoding video, and FFmpeg's x86 assembly hand-optimization runs 62 times faster than equivalent C. The segment also previews the CIA-VLC spy story, the intelligence-agency backdoor request Kempf flatly refused, and Kieran's "no regrets" Twitter philosophy. > *"We care about excellent code. We don't care who you are. Like maybe you're a dog. I don't care, right? I need to look at your code."* — Jean-Baptiste Kempf ## [02:17] Introduction Lex sets the scene: FFmpeg is the invisible backbone behind YouTube, Netflix, Chrome, VLC, Discord, and nearly every platform that touches video or audio. VLC has been downloaded more than 6.5 billion times. Both projects are built entirely by volunteers. Lex frames the episode not merely as a technical discussion but as a tribute to engineers who work for the craft rather than for fame or money—"one of the great examples of human beings quietly collaborating across borders to build something useful, durable, and elegant." > *"It is one of the most incredible software systems ever developed, and it's all done by volunteers."* — Lex Fridman ## [05:35] Weirdest things VLC opens The conversation lightens up with playful examples of VLC's legendary tolerance for exotic formats. Kempf describes users capturing VHS tapes via capture cards, support for DVD-Audio with custom encryption, and the Lucasfilm Star Wars game codec that FFmpeg implemented for a single 10-second opening sequence. At a VideoLAN conference, a competition to create the most broken file ever—an MKV where every frame changed resolution, aspect ratio, and rotation—ended with VLC playing it perfectly. The orange traffic-cone logo is discussed: so recognizable that 25% of VLC's website traffic arrives from people searching "cone player." > *"There was a file that's a valid ZIP and a valid MP3 at the same time or something like that—and VLC opened all of the stupid files."* — Kieran Kunhya ## [09:59] How video playback works Kempf and Kunhya walk through what happens the moment you press play: the player fetches a byte stream from a URL, the demuxer separates audio, video, and subtitle tracks, entropy decoding removes mathematical compression, intra prediction reconstructs still-image frames (I-frames), motion-compensation handles temporal redundancy (P- and B-frames), and the final raw pixels are handed to the GPU or audio card. Video compression achieves 100x to 200x reduction by exploiting how human eyes perceive luminance versus color—working in YUV space rather than RGB—and by reusing unchanged background regions across frames. Kunhya warns that every single sentence in this pipeline represents someone's lifetime of work. > *"Everything we've just said in the past couple of minutes, every sentence is someone's lifetime's work. There are books about every sentence."* — Kieran Kunhya ## [19:20] Video codecs and containers The hosts clarify the often-confused distinction between containers and codecs. A container (MP4, MKV, MOV) multiplexes audio, video, and subtitle tracks; the codec (H.264, AV1) compresses the content inside. VLC and FFmpeg deliberately ignore the file extension and probe the actual bytes—because in the real world, extensions lie. The segment covers how AVI was Microsoft's format, MOV became MP4 via Apple, and the Matroska/MKV format emerged from the open-source community. Modern codecs like AV1 are not single algorithms but collections of tools that adapt to content type—screen share, animation, live video—each requiring different coding strategies. > *"We discard the file format. We look into the file to understand what's in it because so many people say, 'Oh, it's a video, it must be MP4,' but technically it's an MOV or maybe it's a MKV."* — Jean-Baptiste Kempf ## [30:07] FFmpeg explained FFmpeg is described as a low-level library suite—libavcodec, libavformat, libavfilter—plus a command-line tool so expressive that Kempf calls it a full programming language. Every person watching a YouTube video, recording with OBS, or editing in a professional broadcast box is likely touching FFmpeg. Kunhya notes that trillion-dollar corporations and grandmothers with home videos operate on exactly the same technology stack. The segment dives into open-source licensing—MIT, GPL, LGPL, AGPL—as "social contracts" that define community norms. Kempf recounts the painstaking process of re-licensing VLC's core from GPL to LGPL, requiring him to track down more than 350 contributors, including visiting the factory-worker father of a deceased contributor to obtain permission for two lines of code. > *"From a philosophical level, it's incredible that your grandmother's home videos and trillion-dollar corporations are on a level playing field using the same technology stack."* — Kieran Kunhya ## [51:07] Linus Torvalds Kempf offers a nuanced defense of Linus Torvalds's legendary harshness. The Linux kernel's core community is tiny—as is FFmpeg's (10–15 active maintainers)—and those few people must maintain every line of code forever. "We cannot compromise on quality because the core community of FFmpeg is ten to fifteen, and we are the ones who are going to maintain your code." Kunhya adds that terseness is often simply fatigue: volunteers arrive home after a full day of work and review patches without the bandwidth to hand-hold. Kempf also points out that most community members are non-native English speakers, and cultural misreadings amplify perceived hostility. > *"We cannot compromise on quality because the core community of FFmpeg is ten to fifteen, and we are the ones who are going to maintain your code."* — Jean-Baptiste Kempf ## [55:46] Turning down millions to keep VLC ad-free Kempf traces VLC's unlikely origin: a French engineering school (École Centrale Paris) whose student-run campus built a satellite video-streaming system in 1995—a decade before YouTube—just to enable faster networks for video games. From that Network 2000 project grew VideoLAN, and VLC emerged as its client. Kempf joined in 2003 when the project had nearly died, grew it from hundreds of thousands to billions of installs, and along the way repeatedly refused "obscene" offers to bundle toolbars, change search engines, or insert advertisements. His reasoning: "I need to go to bed at night and be happy about what I've done. If I had sold out, I would have betrayed so many other people who work here." > *"I refuse dozens of millions of dollars, yes, several times. Yes, I could be a multimillionaire and be somewhere on the beach. But I did not do it because I thought it was not moral and it was not the right thing to do."* — Jean-Baptiste Kempf ## [70:04] FFmpeg & Google drama Kunhya recounts a public controversy in which Google's security team used AI to auto-generate bug reports for FFmpeg, filing them under tight 90-day deadlines—with some vulnerability reports going to the press before patches could be written—without contributing corresponding fixes or meaningful funding. Kunhya compares it to "a denial of service by AI-generated bug reports" on obscure 1990s game codecs. The saga escalated via spicy FFmpeg tweets (a "rap battle" in Kunhya's words), but produced concrete results: Google began sending patches and established a financial reward system for fixes. A parallel incident saw Microsoft Teams engineers file a high-priority bug on the volunteer tracker, name-dropping their product's scale, and offering a one-time payment of a few thousand dollars in response to a request for a long-term support contract. > *"Google uses FFmpeg at a scale probably you or I couldn't even contemplate—millions of CPU cores. And yes, they contribute in areas mostly regarding their own products. But in a wider sense, there's a disproportionate level of contribution."* — Kieran Kunhya ## [89:18] FFmpeg developers What motivates FFmpeg's volunteer engineers? Kempf identifies three drivers: passion for the subject matter (many contributors arrived because they loved anime), excellence of the craft ("this is the best school ever of programming"), and pride in impact ("you can tell your grandma: I do this so you can play video on your laptop"). Kunhya adds that Andrew Kelley, creator of the Zig programming language, was an FFmpeg developer who credits his time there as his real-world education. Teenagers have written thousands of lines of hand-optimized assembly for FFmpeg. Kieran's favorite quote, from John Collison: "The world is a museum of passion projects." > *"If you're good in C, if you know how to write assembly in FFmpeg, I assure you you're going to be one of the best programmers ever—even if you're working on writing TypeScript."* — Jean-Baptiste Kempf ## [95:55] VLC and FFmpeg Kunhya frames the FFmpeg-VLC relationship as a "binary star system": VLC is to FFmpeg as Android is to Linux—they depend on each other and succeed because of each other. Roughly 80% of FFmpeg pipelines depend on at least one VideoLAN project (most often x264). VLC gives FFmpeg exposure to a vast zoo of real-world broken files. When compiled for Windows, VLC links against about 16 million lines of code, of which only 1 million live in the VLC repository itself. The two projects share many developers and collectively demonstrate that complex software ecosystems can be built entirely from interdependent open-source components. > *"VLC is to FFmpeg as Android is to Linux. They depend on each other, but they coexist because of each other."* — Kieran Kunhya ## [100:29] History of FFmpeg The "eras tour" of FFmpeg begins with Fabrice Bellard creating the initial concept, followed by the Michael Niedermayer era of the early 2000s—exhaustive support for DivX, Xvid, Windows Media, and RealMedia, eliminating the need for bloated, spyware-ridden codec packs. The late 2000s brought H.264 maturity and the rise of high-definition video. Throughout, VLC served as FFmpeg's field test: millions of users exposing edge cases that no lab could anticipate. > *"At the time you needed a new player to play every different type of file format. Having a single library that was fast and open source—that was a massive achievement."* — Kieran Kunhya ## [103:46] Reverse engineering codecs The segment showcases the art of reverse engineering proprietary codecs. Kostya Shishkov—described as "borderline genius"—reverse-engineered 20–30 megabyte binary blobs (each megabyte representing roughly a month of normal work) for fun, producing decoders for Windows Media, RealMedia, and GoToMeeting formats. Kunhya explains the methodology: hook into the proprietary player to dump raw YUV data, open a disassembler, step through machine code instruction by instruction to infer the entropy coding, prediction, and IDCT stages, then validate bit-exactness against sample files. For months, the work produces no visible output—pure debugging in memory. > *"He looked at the world as a binary specification. He didn't need documentation or anything. He would go away and come back and do interesting stuff."* — Kieran Kunhya ## [117:01] FFmpeg testing FFmpeg's FATE (FFmpeg Automated Testing Environment) system runs a pivot table of test combinations: dozens of compilers (GCC, Clang, MSVC, Apple Clang, Intel Compiler), operating systems (Linux, macOS, Windows, BSD, Solaris), and CPU architectures (x86, ARM, RISC-V, PowerPC). All test machines are volunteer-hosted. The system catches compiler miscompilations—rare but devastating, since even a single wrong bit in a frame dependency chain can cascade into major visual corruption. Kunhya notes that the Macs at the top of the FATE dashboard are hosted in his own office. > *"It's not just a matrix at this point. It's like a pivot table of different combinations—all run by volunteers."* — Kieran Kunhya ## [121:08] Assembly code (handwritten) This extended chapter is the technical heart of the episode. Handwritten x86/ARM SIMD assembly in FFmpeg and x264 runs up to 62 times faster than equivalent C—a gap that modern compilers and auto-vectorization cannot close despite years of trying. VLC still supports Windows XP through Windows 11, macOS 10.7 through macOS 26, iOS 9 through the latest, BSD, Solaris, and even OS/2. Understanding assembly forces programmers to internalize CPU pipeline stages, SIMD registers, L1/L2/L3 cache, and memory bus constraints. Kempf and Kunhya introduce the x86inc framework built by Loren Merritt for x264 and JB's Assembly Lessons tutorial series, which have attracted contributions from teenagers learning directly from the source. > *"I believe it's necessary to understand assembly language, even if you don't do it much, to understand what's going on inside your computer. That will make you a better programmer."* — Jean-Baptiste Kempf ## [145:26] Rust programming language Kempf and Kunhya hold divergent opinions on Rust. Kunhya respects the memory-safety goal but finds the community self-important—"It has a very big Esperanto vibe"—and argues that Rust rewrites reaching only 85–90% of required feature coverage are insufficient; "the last 1% takes 99% of the time." Kempf has written Rust VLC modules and sees genuine value, but notes that the lack of training data for low-level SIMD work means AI tools cannot yet assist meaningfully. The discussion broadens to the two assembly wizards of the community: Henrik Gramner, whose knowledge of Intel x86 cycle counts exceeds Intel's own engineers, and Martin Storsjö, who writes ARM Neon assembly on a virtual keyboard while watching his kids play in the playground. > *"Rust reminds me of the Sinclair C5. In order to get people to move, you have to build something as good as, if not better than, what you have now."* — Kieran Kunhya ## [154:42] FFmpeg and Libav fork In 2011, FFmpeg split into FFmpeg and Libav, primarily over governance and leadership style rather than technical disagreements. Several Linux distributions temporarily shipped Libav instead of FFmpeg. Kempf describes open-source forks as healthy—they force projects to confront structural weaknesses. Eventually most of Libav's developers returned to FFmpeg, and the projects merged back. Kempf draws a parallel to the XZ Utils attack, where a lone maintainer, exhausted by coordinated social engineering, granted commit access to an attacker—highlighting how burnout creates the very single-point-of-failure vulnerabilities that make critical open-source infrastructure fragile. > *"Forks are important because they change the status quo of a community. FFmpeg today is better than it was before the fork."* — Jean-Baptiste Kempf ## [163:04] Open source burnout Kempf and Kunhya confront the mental health crisis among open-source maintainers. Kempf has received physical death threats—including a letter containing powder—over decisions such as dropping PowerPC support. The security community's habit of filing alarming CVEs for hobby-project edge cases adds psychological load without providing patches. Kempf now maintains several libraries whose original maintainers burned out. The conversation broadens to the systemic problem: critical infrastructure like libxml and XZ is maintained by one or two people, unknown to the trillion-dollar enterprises that depend on them. > *"The mental health of the open source maintainers is something that large corporations don't care or don't see."* — Jean-Baptiste Kempf ## [170:51] x264 and internet video H.264 transformed internet video by arriving exactly when Intel Core 2/Nehalem CPUs made real-time software decoding practical. The key innovation of x264 was psychovisual rate-distortion optimization—encoding decisions driven by visual quality metrics rather than mean squared error, producing sharper, more natural-looking video. This was driven by the anime community's high standards for perceived sharpness. AV1 offers 40–60% bandwidth savings over H.264 at the same quality, but encoding costs two orders of magnitude more CPU. YouTube therefore re-encodes only popular videos in AV1, making the extra compute worthwhile by amortizing it over millions of viewers. > *"Thirty percent of the video from Netflix is now in AV1, fifty percent of YouTube."* — Jean-Baptiste Kempf ## [184:07] Video compression basics The chapter clarifies I/P/B frame structure: I-frames are complete still images, P-frames reference only previous frames, and B-frames can reference both past and future frames. ProRes is an intra-only codec designed for nonlinear editing—no temporal dependencies, fast seeking. The segment also covers constant-bitrate versus constant-quality encoding, group-of-pictures length, and the thousands of engineers at Netflix, YouTube, and Meta whose entire job is tuning FFmpeg parameters for specific content types. A historical curiosity: Google Video originally used VLC as an ActiveX plugin inside Internet Explorer; today VLC is compiled to WebAssembly to run inside browser JavaScript engines. > *"You have I-frames that are complete frames, P-frames that depend only on I-frames, and B-frames that can depend on frames in front."* — Jean-Baptiste Kempf ## [191:04] CIA and fake VLC WikiLeaks' Vault 7 release revealed that the CIA built a modified version of VLC with an additional DLL (psapi.dll) that silently encrypted and exfiltrated documents while the victim watched a movie, using the expected high CPU load of video playback as cover. VideoLAN issued a press release directing users to download only from the official website. A separate incident involved Chinese state hackers distributing a fake VLC using legitimate signed VideoLAN DLLs to target Indian users, causing India to ban VLC until Kempf fought a successful legal battle to reverse the ban. The segment also surfaces a hidden feature: VLC can render movies as ASCII art in a terminal, useful for diagnosing multicast network paths via SSH. > *"If we had to compromise our software, we would shut it down. This is clear."* — Jean-Baptiste Kempf ## [201:39] Ultra low latency streaming Kempf explains adaptive streaming (HLS, DASH): the player downloads segments, times the download, and adjusts quality tier accordingly. The real engineering frontier is live broadcasting with strict CBR constraints—satellite uplinks cannot burst even for one second. Kempf describes his company Kyber, an open-source (AGPL dual-licensed) ultra-low-latency streaming stack targeting robotics and XR, streaming compressed video feeds to devices without onboard compute. The segment ends with a discussion of teleop for robots, where latency directly determines safety. > *"Kyber is open source. Everything on Kyber is open source. If you want to use it in your product and not open source it, you pay the commercial license."* — Jean-Baptiste Kempf ## [219:07] AV2 codec and video patents AV2, the successor to AV1 within the Alliance for Open Media (of which VideoLAN is a member), promises a further 30% bandwidth reduction. VideoLAN's dav1d decoder will be followed by "dav2d." The Alliance exists specifically to escape the HEVC/H.265 patent thicket: HEVC's three separate patent pools demanded fees so large that HP removed HEVC support from new laptops, and streaming giants calculated they could build a new royalty-free codec for less than the annual licensing cost. France's rejection of software patents means Kempf has never paid codec licensing fees—if he had to, the bill would exceed 200 euros per user. > *"At a hundred million per year, you know, I could create my own codec—and this is what they did."* — Jean-Baptiste Kempf ## [228:59] VLC backdoors Intelligence agencies from two different countries approached Kempf asking him to insert backdoors into VLC. He declined both, in terms he describes as "a lot less polite" than a simple no. The chapter broadens into a discussion of European entrepreneurship: Kempf argues that French startup culture has transformed over 15 years—failure stigma has fallen, AI companies are proliferating—while acknowledging that over-regulation remains a real drag. He closes by reflecting on his strategy for remaining calm under legal and political pressure: always ask "am I dying? Am I hurting someone?" If not, move on. > *"If we had to compromise our software, we would shut it down. Also because what we do is good and it's done for everyone."* — Jean-Baptiste Kempf ## [239:14] Video archiving Kieran profiles the archiving preservation community, led in part by Dave Rice of CUNY, which relies on FFmpeg as a "Rosetta Stone" for playing future-proof multimedia. The community funded FFV1, FFmpeg's lossless codec, to guarantee that archived footage loses no information—critical because lossy compression could destroy forensic or historical details visible only on close inspection. A famous cautionary tale: the BBC's 1986 New Domesday Book project archived content on BBC Micros, and within 20 years no one had working software to read it. There are now more historical video tapes in archives than functional tape heads in the world to digitize them, forcing painful triage decisions about what human history to preserve. > *"C will be like Latin. It will be a thing you learn from the past, but it will still be usable in certain contexts."* — Kieran Kunhya ## [245:51] Future of FFmpeg and VLC The closing chapter surveys where multimedia is heading: volumetric video, point-cloud codecs for robotics, RGBD depth streams, XR/VR streaming, and—speculatively—neural interfaces that may one day require codecs for compressed brain data. Kempf is confident FFmpeg will exist in 100 years; VLC he rates as "maybe." He closes with his personal philosophy: "Regrets are a tax on your mind. Learn from your mistakes, but don't regret." The episode ends with Lex reading Linus Torvalds: "Most good programmers do programming not because they expect to get paid or get adulation by the public, but because it is fun to program." > *"Regrets are a tax on your mind. Learn from your mistakes, but don't regret. Because you've done it, so unless you have a time machine, don't regret."* — Jean-Baptiste Kempf ## Entities - **Jean-Baptiste Kempf** (Person): President of VideoLAN, primary maintainer of VLC, founder of Kyber and several other companies; declined tens of millions of dollars to keep VLC ad-free. - **Kieran Kunhya** (Person): Veteran FFmpeg contributor, codec engineer, founder of Open Broadcast Systems, the voice behind the FFmpeg account on X. - **Lex Fridman** (Person): Host of the Lex Fridman Podcast, AI researcher, longtime VLC and FFmpeg advocate. - **Fabrice Bellard** (Person): Creator of FFmpeg, QEMU, and tcc; foundational figure of the project. - **Michael Niedermayer** (Person): Long-time FFmpeg maintainer who drove exhaustive codec support through the 2000s. - **Kostya Shishkov** (Person): Legendary FFmpeg reverse engineer who decoded proprietary binary blobs for Windows Media, RealMedia, and GoToMeeting codecs. - **Henrik Gramner** (Person): Assembly wizard with deeper knowledge of Intel x86 cycle counts than Intel's own engineers. - **Linus Torvalds** (Person): Creator of Linux and Git; referenced as a model of uncompromising code quality standards in open-source communities. - **FFmpeg** (Software): Open-source multimedia framework providing codecs, muxers, filters, and command-line tools; the invisible backbone of nearly all internet video. - **VLC** (Software): Open-source media player with 6.5+ billion downloads, built on libVLC and FFmpeg; plays virtually any format on any platform. - **x264** (Software): VideoLAN's open-source H.264 encoder; the dominant software encoder for internet video, famous for psychovisual optimizations. - **dav1d** (Software): VideoLAN's fast open-source AV1 decoder; widely deployed in browsers and streaming clients. - **VideoLAN** (Organization): French nonprofit that stewards VLC, x264, dav1d, and related open-source multimedia libraries. - **Alliance for Open Media** (Organization): Industry consortium including Google, Netflix, Apple, Amazon, and VideoLAN that created AV1 and is developing AV2 as royalty-free codec standards. - **FATE** (Software): FFmpeg Automated Testing Environment; volunteer-hosted CI grid testing hundreds of compiler/OS/architecture combinations. - **Kyber** (Organization): JB Kempf's startup building an ultra-low-latency open-source streaming stack for robotics and XR, dual-licensed AGPL/commercial. - **H.264 / AVC** (Concept): The dominant internet video codec standard; open-source implementation is x264; basis of Blu-ray and most MP4 files. - **AV1 / AV2** (Concept): Royalty-free next-generation video codec standards from the Alliance for Open Media; AV1 saves 40-60% bandwidth vs H.264; AV2 adds another 30%.

La présentation officielle d'Anthropic sur Claude Code — ce que c'est, en quoi il diffère de Claude.ai, et les trois choses à savoir avant de laisser un LLM exécuter des commandes sur votre base de code. Destiné aux développeurs sur le point d'installer l'outil terminal pour la première fois. ## [00:04] Ce qu'est Claude Code et où il s'exécute Claude Code se positionne comme un outil de codage agentique : il comprend votre base de code, édite des fichiers, exécute des commandes et s'intègre aux outils de développement que vous utilisez déjà. Il fonctionne sur plusieurs surfaces — terminal, VS Code, JetBrains IDE, l'application de bureau Claude et le web — mais cette présentation se concentre sur le terminal comme expérience de référence. > *Claude Code is an agentic coding tool that understands your code base, edits your files, run commands, and integrates with your existing developer tools to help you get things done faster.* ## [00:34] En quoi il diffère de Claude.ai La distinction clé n'est pas la capacité du modèle mais l'accès : Claude Code accède directement à votre terminal et à l'intégralité de votre base de code, ce qui élimine le cycle copier-coller dans le chat — l'outil effectue le travail sur place. L'appeler "agent IA" résume cette surface d'exécution directe. > *Unlike Claude AI, Claude Code has direct access to your files in your terminal and your entire code base.* ## [00:51] Les agents IA et ce que Claude Code peut faire Un agent IA désigne ici un logiciel qui interagit avec son environnement et prend des mesures pour atteindre un objectif défini — dans sa forme la plus basique, un LLM en boucle en temps réel avec accès à des outils, des services externes et d'autres agents. Pour Claude Code, cela se traduit par des capacités concrètes : lire et expliquer votre base de code, tracer des bugs à travers les fichiers, exécuter des scripts de build et des tests, installer des paquets, et récupérer sur le web la documentation API la plus récente pour décider de la prochaine action. > *An AI agent is a software that can interact with its environment and perform actions to complete a defined goal.* ## [01:45] Trois concepts à connaître avant de commencer Le narrateur signale trois propriétés qui façonnent l'utilisation quotidienne. Premièrement, la **fenêtre de contexte** est la mémoire de travail de Claude — grande mais finie — c'est pourquoi l'agent doit naviguer stratégiquement dans une base de code plutôt que de tout charger. Deuxièmement, Claude Code **demande une autorisation** avant d'exécuter des commandes ou de modifier des fichiers ; vous gardez le contrôle que vous souhaitiez piloter chaque étape ou le laisser fonctionner de manière autonome. Troisièmement, **il peut se tromper** : mal interpréter l'intention, introduire des bugs ou sur-ingéniérer une correction. Traitez les sorties comme vous le feriez avec n'importe quel outil, sans les considérer comme parole d'évangile. > *By default, Claude Code will ask you before running commands or making changes to your code base.* ## [02:34] Récapitulatif Claude Code est un outil de codage agentique qui lit votre base de code, édite des fichiers, exécute des commandes et se connecte à des outils externes pour vous aider à livrer plus vite — disponible dès maintenant sur le terminal, VS Code, JetBrains et l'application de bureau Claude. > *Claude Code is an agentic coding tool. It reads your code base, edits your files, runs commands, and connects to external tools to help you ship faster.* ## Entités - **Anthropic Tutorial Narrator** (Person): Le narrateur officiel d'Anthropic pour la série de tutoriels Claude Code 101. - **Claude Code** (Software): L'assistant de codage agentique basé sur le terminal d'Anthropic, qui opère directement sur votre base de code. - **Claude.ai** (Software): Le produit Claude basé sur le chat — mis en contraste avec l'exécution en environnement de Claude Code. - **AI agent** (Concept): Un LLM fonctionnant en boucle en temps réel avec accès à des outils, des services externes et d'autres agents pour poursuivre un objectif défini. - **Context window** (Concept): La mémoire de travail de Claude — finie, c'est pourquoi l'agent navigue stratégiquement plutôt que de charger la base de code entière. - **VS Code / JetBrains IDEs** (Software): Les intégrations d'éditeurs dans lesquelles Claude Code est disponible, aux côtés du terminal et de l'application de bureau Claude.

Alex Lupsasca — 2024 New Horizons Breakthrough Prize winner and OpenAI resident scientist — recounts how GPT-5 resolved a year-long open problem in quantum field theory: proving that single-minus gluon tree amplitudes are non-zero and finding their compact closed form. He then describes how the publicly available GPT Pro, given the gluon paper as a seed, independently generalized the result to graviton amplitudes in under three days of human clock time. Throughout the conversation, Lupsasca reflects on what this trajectory means for how physics is done, how the next generation of physicists will be trained, and where the remaining bottlenecks — verification, creativity, and publishing infrastructure — still lie. ## [00:00] Introduction to AI's impact on physics research Lupsasca opens in medias res, framing the episode's central claim before the formal introduction: AI has crossed a threshold where it can resolve questions that stumped human experts for over a year. He describes this not as a curiosity for theoretical physicists but as a profound, if underappreciated, change in the nature of scientific discovery itself. > *"That's a certain milestone that we've passed, and I think maybe for the average person on the street who doesn't care about theoretical physics, this is not very noticeable, but I think it's a very profound change and we've really passed some kind of a threshold."* ## [00:43] Guest introduction: Alex Luposka The hosts — Brandon (Atomic AI) and RJ Honicky (Miro Omix) — introduce Lupsasca as a Vanderbilt professor and OpenAI fellow who holds both the 2024 New Horizons in Physics Breakthrough Prize (often called the "Oscars for science") and the IUPAP Young Scientist Award. Lupsasca immediately sets the narrative arc: a year ago, AI was useful for email but not for his work; ChatGPT o3 was the first model that genuinely helped with research math; then GPT-5 reproduced one of his hardest published results in 30 minutes. > *"When GPT-5 came out it was able to reproduce one of my best papers that took me a very long time to come up with in like 30 minutes. And that's when I really became AI pilled."* ## [02:49] Alex joining OpenAI and the shift in physics research After GPT-5's release, Lupsasca began evangelizing the shift to colleagues who were skeptical. Finding OpenAI equally excited, and being on sabbatical, he joined as resident scientist — the person physicists around the world now email when something astonishing happens. He describes receiving an inbound that week about Codex simulating the Sachdev-Ye-Kitaev (SYK) model in 10 minutes, a feat that many research groups had struggled to achieve due to the narrow Venn diagram of physicists with strong coding skills. > *"I talked to OpenAI. They were also really excited and I thought I have to get in on this and to understand that this is happening and not be a part of it is a huge mistake so I have to go to OpenAI."* ## [04:08] The release of GPT-5 and the shift in capabilities Lupsasca contrasts the lukewarm Twitter reception of GPT-5 (complaints that it was not better at writing email) with what he observed at the science frontier. He notes GPT-5.4 is another significant jump, and describes how AI capabilities for physics have been accelerating rapidly since o3, the first reasoning model strong enough for research-grade mathematics. He uses this as a bridge to the central technical story of the episode: a pair of new papers on gluon and graviton scattering amplitudes. > *"At the science frontier the capabilities were really taking off."* ## [10:05] Explaining Quantum Field Theory and amplitude calculations Lupsasca gives an accessible primer on quantum field theory (QFT), the framework that reconciles special relativity and quantum mechanics. The key objects in QFT are scattering amplitudes — complex-valued functions that encode the quantum probability for a set of incoming particles (with given energies, momenta, and polarizations) to scatter into a set of outgoing particles. These amplitudes are computed at particle colliders like the LHC, and knowing the n-point amplitude (for any number n of particles) encodes essentially the full content of the theory. > *"If you have a particular force and you're able to compute the n-point amplitudes... you know everything about the theory."* ## [14:20] Overview of gluons and the strong force Gluons are the force-carrying particles of the strong nuclear force — the force that, despite like-charge repulsion between protons, holds the atomic nucleus together. They are the QFT analog of photons for electromagnetism and gravitons for gravity. Like photons, gluons carry a polarization (helicity): positive (right-handed) or negative (left-handed). This helicity structure is central to the paper discussed next. > *"The strong force is mediated by the exchange of the particles of the strong force, which are called gluons, because they're what glues together the nucleus of the atom."* ## [14:38] Discussing the first research paper on single-minus gluon tree amplitudes Lupsasca unpacks the paper's title — "Single-Minus Gluon Tree Amplitudes Are Non-Zero" — piece by piece. Tree amplitudes are the leading-order (no-loop) contributions to scattering. All-plus-helicity amplitudes are exactly zero by a symmetry argument. Single-minus amplitudes — where all but one gluon have positive helicity — were assumed in textbooks to also be zero by the same argument. The paper proves they are not. The result involves collaboration with Alfredo Guevara (IAS), David Skinner (Cambridge), Andrew Strominger (Harvard), and Kevin Wheel. > *"If you look at the lecture notes and textbooks that have been written on this, the same argument that rules out the all-plus amplitudes also appears to rule out the single-minus amplitudes."* ## [20:56] How ChatGPT helped solve a year-long physics puzzle Strominger, Guevara, and Skinner had understood for about a year that the textbook argument has a loophole: when particles are collinear (exactly aligned in momentum), the standard dimensional-analysis reasoning fails, and single-minus amplitudes can be non-zero. But computing what those non-zero amplitudes equal had eluded them. Lupsasca invited Strominger to visit OpenAI and work on it with AI. The week before Strominger's flight, Lupsasca began using ChatGPT Pro. By the time Strominger landed, they had the answer. > *"Using ChatGPT we solved the problem before he even got off the plane."* ## [23:02] Complexity of manual calculations in physics Lupsasca shows the audience a concrete illustration of the difficulty: the six-point single-minus amplitude, worked out by hand by Alfredo Guevara, is a sum of 32 terms each of which is itself a product of four complicated factors. The number of terms grows factorially with the number of particles n — super-exponential growth. This is the messy representation that the group had been staring at for a year, seeking the analog of the elegant Parke-Taylor formula. > *"By the time you get to six terms, it explodes in your face."* ## [26:12] The history and mechanics of Feynman diagrams Feynman diagrams are a visual language introduced by Richard Feynman to organize perturbative QFT calculations: diagrams represent possible intermediate histories of a scattering process, and the full amplitude is a sum over all of them. Diagrams are organized by number of vertices (interaction points); each additional vertex is suppressed by the coupling constant, so tree diagrams (fewest vertices) dominate. Loop diagrams — where intermediate particles are created and annihilated — contribute smaller corrections. The combinatorial explosion of tree diagrams is the root cause of factorial growth. > *"In principle, there are infinitely many pictures to sum over."* ## [27:44] The Parke-Taylor formula and the quest for simplification In the 1980s, Parke and Taylor computed the "maximally helicity violating" (MHV, or double-minus) gluon amplitudes through a heroic Feynman diagram expansion. Despite the factorial number of terms, everything canceled to leave a single compact formula — the Parke-Taylor formula — that fits in half a line. Strominger, Guevara, and Skinner spent a year looking for the analogous compact formula for the single-minus case. Their search stalled at the level of the messy Feynman representation. > *"Andy, Alfredo and David spent the last year chasing the analog of the Parke-Taylor formula, the very simple answer that was obtained in the '80s for the double minus amplitudes."* ## [31:26] Using ChatGPT to find the simplification in the special phase space region When the five-point single-minus amplitude was fed to ChatGPT Pro, the model identified a special subregion of phase space (where one particle's frequency has opposite sign) in which the amplitude simplifies from eight terms to a product of just three. This appears not to have been a known fact; the model wrote Python code and tested thousands of possibilities to deduce it. Moving to the six-point amplitude (Guevara's hand calculation), ChatGPT simplified 32 terms to a product of 4. It then conjectured the general n-point formula — with only linear growth in the number of terms, the best possible behavior. GPT-5.2 Pro guessed the formula but could not prove it. > *"The formula that it proposed, instead of having this factorial growth... here it's actually linear. So if you double the number of particles, you only double the number of terms."* ## [38:07] Proving the formula from scratch to ensure validity To obtain a proof, Lupsasca used an internal OpenAI model with extended reasoning. He gave it the problem cold — without the conjectured formula — and asked it to find the general answer in the special phase-space region. After 12 hours of computation, the model independently rediscovered the same formula and produced a complete three-step proof. The proof constitutes the bulk of the published paper. The team kept the AI attribution to one paragraph, framing the paper as a physics result that stands on its own merits. > *"We gave it the whole problem from scratch... and it came back with the same formula which we had not given it. So it rediscovered the correct formula. But this time it also found the proof."* ## [41:00] Determining the scientific impact and future research Asked to compare the result to the Parke-Taylor formula, Lupsasca is candid that scientific impact is only assessable decades later, but argues the result is genuinely surprising and should open a line of attack toward deeper questions in quantum gravity. The conversation pivots naturally to the second paper. > *"I think the true value of a paper can only be assessed decades into the future based on how much future work it leads to and what developments it opens up."* ## [42:27] Introduction to the second paper on graviton amplitudes Gravitons are the hypothetical quanta of gravity — the spin-2 force carrier analogous to the spin-1 photon (electromagnetism) and gluon (strong force). Unlike gluons, gravitons have never been directly detected, but they are central to quantum gravity theory. The second paper, "Single-Minus Graviton Tree Amplitudes Are Non-Zero," shows the same loophole applies to gravity and that a compact formula extends there too — despite gravitons being mathematically more complex than gluons. > *"We wrote this paper which is called single minus graviton tree amplitudes are non-zero. So it's the same title almost, except with graviton instead of gluon."* ## [45:41] Defining particles, irreducible representations, and symmetry Lupsasca sketches the modern QFT definition of a particle (an irreducible representation of the Poincaré group, classified by Wigner according to mass, spin, and charge) and explains why gravitons are spin-2 while gluons and photons are spin-1, making graviton polarization data twice as rich. Crucially, the second paper was complete within three days of the first going public — most elapsed time was spent verifying correctness, not computing. > *"Most of the time was spent verifying the answer, not writing, which is insane, actually, if you take a step back."* ## [47:46] How GPT Pro generalized the research to gravity For the graviton paper, no internal model was needed — the publicly available ChatGPT GPT-5.2 Pro sufficed. Lupsasca provided the gluon paper as context plus two paragraphs describing the key mathematical changes, then said "Good luck. You're a brilliant theoretical physicist." Over a 110-page exchange, the model worked through the graviton calculation — applying the directed matrix tree theorem, a piece of known combinatorics that neither Lupsasca nor collaborators had thought to invoke — produced correct intermediate results, and wrote a draft paper very close to the final arXiv version from section 3 onward. > *"It's a real solid result in quantum gravity that was done pretty much completely by an AI with human steering it and asking kind of the right questions."* ## [53:57] The epistemological shift: Is this a new way of doing physics? The hosts raise the central epistemological question: if an undergraduate with domain knowledge and good prompting could have done this, what does graduate training mean now? Lupsasca agrees this is the hardest open question facing academia. He notes that arduous calculation trains not just skill but self-confidence, that the gap between coursework and the research frontier is growing, and that many "easy" problems professors once assigned to students are now solvable by AI in minutes. He offers two concrete ways AI has already changed his own workflow: dramatically reducing time spent confused between steps, and enabling parallel AI scouts that explore multiple research directions simultaneously. > *"With AI, actually, you can launch 10 instances of chat and have each one try a different route and send it as a scout that moves very fast into the unknown."* ## [59:27] The use of AI as a 'scout' for research directions Lupsasca elaborates on the scout metaphor: rather than carefully mapping a route from A to C before committing to it, a researcher can now dispatch many AI "scouts" in parallel, get rapid feedback on which directions are promising, and redirect human attention accordingly. Even when a scout makes errors, its signposts reduce orientation cost for the human following. This constitutes a qualitatively new mode of research — one where the bottleneck shifts from calculation to judgment about which directions matter. > *"Even if ChatGPT doesn't always get everything right, just kind of having a scout that signposts some key steps along the way that you can use to anchor your own movement is extremely helpful."* ## [61:44] The role of 'taste' and collaboration with AI The hosts push on the problem of "taste" — the ability to identify which questions are at the productive edge of knowledge. Lupsasca argues that working effectively with ChatGPT requires the same skill a professor develops advising students: knowing what question to give, at what level of detail. "Taste" — knowing where the frontier is and which questions there are tractable — is the last skill to develop and the one AI currently lacks. AI is, he says, like an extremely technically skilled graduate student: given a sharp, well-posed question, it can do incredibly hard computations correctly, but it does not yet know which question to ask. > *"The difference between a good physicist and a great physicist is knowing what is the right question to ask — that is actually the hardest part of being a scientist."* ## [70:23] Personal evolution from AI skeptic to resident scientist Lupsasca recapitulates his personal arc: skeptic → converted by o3 (which solved in 11 minutes a calculation that would have taken him days) → "AI-pilled" by GPT-5 (which reproduced, in 30 minutes, his best published result on black hole Love numbers and tidal symmetries — a paper whose training cutoff predated its arXiv release) → now resident scientist at OpenAI. He notes that no competing model at the time could match GPT Pro on that calculation. > *"In under 30 minutes, with one hint... it completely solved this problem, which is one of the nicest calculations that I've ever done."* ## [72:46] Solving a black hole perturbation problem with GPT-5 Lupsasca details the "Move 37" moment that converted him: his paper "Why Is There No Love in Black Holes?" establishes new symmetry generators for perturbations of a Kerr black hole (explaining why black hole Love numbers — tidal response coefficients, named after mathematician Augustus Love — are exactly zero). When GPT-5 Pro was first given the full problem cold, it failed. But after being primed with the simpler flat-space warm-up (a 200-year-old known result), it then solved the full Kerr black hole problem in 18 minutes. > *"GPT-5 was able to reproduce one of my hardest calculations, which I think the number of people in the world that could do that you could count on your hands."* ## [76:34] Discussing whether AI can make original, conceptual leaps The hosts ask whether AI is doing genuine recombination versus true creative leaps. Lupsasca cites Terry Tao, who has not yet seen an AI proof that cannot be traced to an obscure reference. But Lupsasca has been impressed and frames the distinction as one of degree rather than kind — humans may also be recombination machines. He believes continued scaling will produce feats of insight that look like creativity, and notes OpenAI is actively working on enabling models to take bigger, more out-of-distribution leaps suited to scientific discovery. > *"I'm not sure there's a qualitative difference. I think it's just a matter of degree — as we continue scaling the capabilities, I don't see why it's going to stop."* ## [80:09] Challenges of 'AI slop' and the future of academic publishing With models now capable of turning out a physics paper in 30 minutes when properly steered, the arXiv preprint server is being flooded with submissions. Lupsasca distinguishes legitimate use (expert steering + careful verification) from "AI slop" — poorly prompted outputs submitted without adequate checking. His proposed response: raise the bar rather than increase volume. The single-minus amplitude papers open a clear line of attack toward genuine quantum gravity questions; the goal should be to pursue harder problems, not to publish incrementally. > *"Instead, I think now that we have this new tool that gives us AI superpowers, I think we should just raise the bar for what it means to write a good paper."* ## [83:13] The bottleneck of writing academic papers Asked what single bottleneck he would remove, Lupsasca nominates the paper-writing process itself — finding it increasingly strange that researchers use AI to do calculations, compress results into a static paper, and then readers feed that paper back into AI to understand it. He envisions interactive, LLM-embedded papers as a plausible future. He also identifies two missing capabilities in current models: (1) the spark of creativity to identify the next important question, and (2) reliable self-verification, so that the onus of checking long AI-generated proofs does not fall entirely on humans. > *"Maybe some kind of interactive paper which lives in some LLM. Maybe your whole paper is some ChatGPT page... I think we're going to head in that direction."* ## [90:19] Final takeaways and looking ahead to the next year Lupsasca's closing message: pay attention. The trajectory from "useful for email" to "solves open problems in quantum gravity" has taken roughly 18 months. Models are solving open problems that expert communities spent years on. Extrapolating forward, with more scaling already in the pipeline, the next 6 to 12 months should bring further surprises. The right posture is excitement, careful verification, and a commitment to pursuing harder problems. > *"If you just extrapolate that into the future, imagine where we're going to be in 6 months or a year — I think it's kind of surreal to live through this time, but it's really happening."* ## Entities - **Alex Lupsasca** (Person): Theoretical physicist, Vanderbilt University professor and OpenAI resident scientist; 2024 New Horizons Breakthrough Prize and IUPAP Young Scientist Award winner; expert in black hole physics and scattering amplitudes. - **Andrew Strominger** (Person): Harvard professor and Lupsasca's former PhD advisor; pioneer of celestial holography; co-author of both single-minus amplitude papers. - **Alfredo Guevara** (Person): Postdoctoral researcher at the Institute for Advanced Study (IAS); performed the foundational hand calculations underpinning the AI-assisted breakthrough. - **David Skinner** (Person): Professor at Cambridge University; co-author of the single-minus gluon amplitude paper. - **Terry Tao** (Person): Fields Medal-winning mathematician at UCLA; referenced regarding the question of whether AI proofs involve genuine creativity. - **Scattering Amplitudes** (Concept): Complex-valued functions in quantum field theory encoding probabilities for particles to scatter; the central mathematical objects of both papers discussed. - **Single-Minus Gluon/Graviton Amplitudes** (Concept): Tree-level scattering amplitudes where all but one particle have positive helicity; previously assumed zero in textbooks but shown non-zero in a collinear phase-space region. - **Parke-Taylor Formula** (Concept): Compact closed-form result for maximally helicity violating (MHV, double-minus) gluon amplitudes derived in the 1980s; the template whose analog was sought for single-minus amplitudes. - **Feynman Diagrams** (Concept): Diagrammatic technique to organize perturbative QFT calculations; individual diagrams represent distinct intermediate-particle histories whose amplitudes are summed. - **Love Numbers** (Concept): Coefficients encoding tidal deformability; famously vanish for black holes, a fact connected to hidden symmetries studied in Lupsasca's "Why Is There No Love in Black Holes?" paper. - **Celestial Holography** (Concept): Research program exploring symmetries of quantum gravity via scattering amplitude structure; motivates studying graviton amplitudes. - **OpenAI** (Organization): AI research company where Lupsasca serves as resident scientist; developer of GPT-5 and the internal extended-reasoning model used for the amplitude proof. - **arXiv** (Organization): Open-access physics and mathematics preprint server; mentioned in the context of AI-generated "slop" flooding submissions. - **GPT-5 / ChatGPT Pro** (Software): OpenAI's frontier language model used as the primary AI tool in both amplitude papers; capable of extended reasoning steps of 20-34 minutes per prompt.

Boris Cherny, the creator of Claude Code, sits down with Sequoia's Lauren Reeder at AI Ascent 2026 and makes a blunt claim: for the code he writes, coding is already solved. He hasn't typed a line by hand in 2026, runs dozens of agent "loops" at once, and ships most of his work from his phone. The throughline is a bet that writing code is becoming so cheap that the interesting questions move up a level — to what teams look like, what happens to software products, and whether the printing press is the right way to think about what's coming. ## [00:00] Introduction A Sequoia emcee opens the AI Ascent session by asking the room for a show of hands: who uses Claude Code, and who has "Claude Code psychosis." She introduces Boris Cherny as the creator of the tool and hands the interview to Sequoia's Lauren Reeder. > *"We know that the entirety of software development kind of rests on your shoulders."* ## [00:55] Claude Code Crowd Check Reeder frames the conversation for a room full of builders and fills in Boris's background: a career writing code, a TypeScript textbook, an engineer's engineer. The detail that lands hardest is recent — as of early 2026 he hasn't written a single line of code by hand, a sharp reversal for someone who built his reputation on craft. > *"Last time we chatted you hadn't written a single line of code in the last year, or at least so far in 2026, which is quite the change."* ## [02:39] Origin Story of Claude Code Boris explains that Claude Code started almost by accident inside Anthropic Labs, a small incubator he joined in late 2024 that also produced MCP and the desktop app. The team built what it wanted, disbanded, and has since reunited under Mike Krieger for a second round. The motivation was a sense of "product overhang" — capability sitting unused because no product had caught up to it yet. > *"The reason that I started to work on coding is we felt like there was this product overhang."* ## [03:35] From Typeahead to Agents In late 2024 the state of the art was typeahead — press tab, complete one line — which Sonnet 3.5 had just made viable. Boris bet the model was nearly ready to skip that step and write all the code as an agent. It didn't work for the first six months; even after release, Claude Code wasn't a hit. The exponential growth only arrived with Opus 4. > *"I built it, and it just really didn't work for the first 6 months. It was barely usable."* ## [05:07] Is Coding Solved Reeder presses on Boris's on-the-record claim that coding is solved. He polls the room — hand-written code versus fully agent-written — and lands the audience around "50% solved," then says for him it's 100%. He points out the Claude Code codebase itself is unglamorous TypeScript and React, chosen deliberately because that stack is heavily represented in the model's training distribution. > *"For me it's just solved."* ## [06:50] Boris Personal Workflow Boris walks through a setup he first shared on Twitter six months ago and didn't expect to surprise anyone. It has since changed: most of his work now happens from his phone, through the Claude app's code tab, where he keeps five to ten sessions each running a few hundred agents. The tool he reaches for most is the loop — fire-and-forget agents that grind on a task and report back. > *"I sort of feel like loops are the future at this point."* ## [08:51] Future Teams and Generalists Asked what teams will look like, Boris predicts a shift toward generalists. Today a generalist still means an engineer who spans iOS, web, and server; tomorrow it means people who are cross-disciplinary, blending engineering with product and design rather than staying in a single lane. He notes the Claude Code team already skews this way. > *"There's going to be a lot more generalists... generalists that are cross-disciplinary."* ## [10:26] SaaS Apocalypse Predictions Reeder asks the question Boris calls his favorite: if AI makes writing code 10 to 100x cheaper, does the value of software products collapse — a SaaS apocalypse? Boris argues the two things that will actually happen aren't the ones people keep predicting, and uses his guest spot on the Acquired podcast as a detour into why he thinks the conventional framing misses the point. > *"I think there's two things that are going to happen and I don't think either of them is the thing that people have been talking about."* ## [12:57] Audience Q&A Deep Dive The floor opens to the room. An audience member, Dan, asks how much of Claude Code's success Boris attributes to the model versus product decisions — Boris says a mix, roughly 50/50, and won't forecast two years out because the team plans a week at a time. The richest answer is his printing-press analogy: before the press, about 10% of Europe was literate; in the 50 years after, more was published than in the prior thousand, and literacy eventually climbed toward 70%. He uses it to argue that building software is on track to become a near-universal skill. Later questions probe the engineering-versus-world gap, local versus cloud models, and how to parallelize agents with loops, batches, and sub-teams. > *"In the 50 years after the first printing press, there was more literature published in Europe than in the thousand years before."* ## [23:35] Closing and Whats Next For the last question, Boris is asked what kind of product he'd build today that gets more interesting as models improve. He points to Claude Design as a good example — decent now, much better soon — and teases features landing for Claude Code in the coming weeks, plus more work on loops, batch, and massively parallel agents, with computer use in the mix. > *"I think loop and batch and things like this around like massively parallelizing agents, that's going to get better."* ## Entities - **Boris Cherny** (Person): Creator of Claude Code at Anthropic; former Anthropic Labs member, now back on the team under Mike Krieger. - **Lauren Reeder** (Person): Sequoia Capital partner; interviewer for this AI Ascent session. - **Mike Krieger** (Person): Chief Product Officer at Anthropic and Instagram co-founder; leads the reunited Claude Code team. - **Anthropic** (Organization): AI lab behind Claude and Claude Code. - **Claude Code** (Software): Anthropic's agentic coding tool, originated in Anthropic Labs alongside MCP and the desktop app. - **Anthropic Labs** (Organization): Internal incubator where Claude Code, MCP, and the desktop app were built. - **Product overhang** (Concept): Model capability that outpaces the products built on it — the gap Boris set out to close. - **The loop** (Concept): Fire-and-forget agent runs that work a task continuously and report back; Boris's most-used workflow. - **SaaS apocalypse** (Concept): The thesis that cheap AI-written code collapses the value of software products — which Boris pushes back on. - **Printing press analogy** (Concept): Boris's frame for AI coding — literacy went from ~10% to ~70% over centuries; software-building may follow.

NYU Stern professor and serial entrepreneur Scott Galloway delivers a two-hour reality check on artificial intelligence: the doom-and-gloom predictions from AI CEOs are largely fundraising theatre, yet the technology poses a genuinely insidious risk that almost nobody is discussing — an epidemic of loneliness. Galloway argues that AI primarily benefits the already-wealthy, that tech leaders should not be trusted to self-regulate, and that the most valuable human skill in the AI era is not coding or Mandarin — it is the ability to endure rejection. The conversation weaves through geopolitics, investing, the masculinity crisis, and what it means to find purpose, closing with a raw reflection on grief and fatherhood. ## [00:00] Intro Host Stephen introduces Scott Galloway against a backdrop of rapid AI development and unsettling quotes from tech CEOs predicting total job replacement. Galloway opens with his central thesis: the two greatest brand collapses of the past 18 months are the United States' global reputation and artificial intelligence itself — both victims of overpromising and poor trust management. He signals that he is an AI optimist at the macro level, but insists the people building it do not have the public's best interests at heart. > *"These techs, they do not have our best interests at heart."* ## [02:45] What's Actually True About AI Galloway reveals a striking data point: approval of AI is directly correlated with income. Only households earning over $200,000 per year hold a net-positive view of the technology, because they benefit through rising portfolios and are the heaviest users. Everyone else sees higher electricity bills, no equity stake in the companies, and dismissive comments from leaders like Sam Altman telling people to stop complaining about energy costs. The AI brand, he argues, has shifted in 18 months from "scary but optimistic" to "scary and only good for the already rich." > *"Your view of AI is directly correlated to your wealth. The only cohort that has a positive rating of AI is people making over $200,000."* ## [05:00] Are AI CEOs Exaggerating The Future To Raise Billions? Galloway lays out the economic logic behind AI catastrophizing. These companies sit on astronomical valuations that can only be justified if either (a) a trillion dollars in incremental revenue materialises from AI-powered products, or (b) there is a massive wave of labour cost savings. Because option (a) is not yet visible — he sees no AI-driven products at meaningful scale — the CEOs amplify option (b), painting vivid pictures of job destruction to justify the efficiency gains enterprises need to believe in. He calls some of the doom talk "thinly veiled fundraising," noting that founders catastrophize and then take secondaries and leave for Santorini. > *"The catastrophizing is nothing more than a thinly veiled attempt to say my technology is so devastating that it's going to shift society and you should invest at this crazy valuation."* ## [09:00] What Would Prove The AI Skeptics Wrong? Asked where he could be wrong, Galloway is specific: if unemployment rises to 20% even temporarily, history shows civil unrest follows regardless of eventual job recovery. He points to radiologists and coders as cases where AI has augmented rather than eliminated roles — new coder job listings are up 11% year-on-year. His benchmark for being wrong is sustained destruction outpacing creation fast enough that the recovery "V" triggers social breakdown before the other side is reached. > *"At 20% unemployment, especially among youth, especially young men tend to get very angry and take to the streets."* ## [11:05] Could AI Move Too Fast For Society To Handle? The conversation turns to pace of change. Galloway uses the host's own media empire — 220 hires in 24 months — as a live counter-example to the apocalypse narrative. He notes a structural inversion: for the first time in decades, unemployment among non-college graduates is lower than among college graduates because AI data centres are driving a boom in trades. He praises the entrepreneurial wave unlocked by AI tools and flags Denmark's 2% GDP commitment to retraining versus America's inadequate equivalent as the real policy failure. > *"AI is not going to take your job. Someone who understands AI is going to take your job."* ## [16:05] What Happens When AI Combines With Robots? Galloway addresses Elon Musk's Optimus robot predictions and the convergence of physical automation with AI cognition. His 2026 stock pick is Amazon, which already holds more industrialised robots than the rest of the US combined and plans to double its retail operation by 2032 without additional headcount. He is sceptical of domestic humanoid robots but takes seriously the military application of weaponised autonomous systems as a genuinely dark unknown frontier. > *"Amazon is saying they're going to double their largest business, which is their retail business by 2032 without an incremental hire using robotics, industrialised robots."* ## [19:05] Is Elon Musk Selling Vision or Reality? Galloway separates Musk the innovator from Musk the stock promoter. He calls Starlink the best tech product of the past several years and credits Musk with inspiring the EV race. But Tesla should trade at 30x earnings, not 150x, and capital will migrate to SpaceX when it IPOs at a projected 90–110x revenues. The core insight: the modern CEO's job has inverted from underpromise-and-overdeliver to overpromise-and-underdel in order to access cheap capital and pull the future forward. > *"The key attribute of an innovator right now is storytelling — to make sure the promise is way ahead of the performance such you can access cheap capital and pull the future forward."* ## [24:05] Which Jobs Are First To Disappear In The AI Shift? Long-haul trucking is Galloway's clearest near-term casualty: autonomous trucks can run the 10 pm to 4 am window and trucking is the largest single employer of non-high-school-graduate males in America. Legal work at the junior associate level is already being displaced — he now routes contracts through two competing LLMs rather than a $400–$2,000 law firm review, projecting a third reduction in his annual legal spend. The pattern he observes is multiplication: one AI-fluent analyst replaces five, yet the resulting EBITDA funds expansion that creates new jobs elsewhere in the ecosystem. > *"AI is not going to take your job. Someone who understands AI is going to take your job. So have a second screen — always have a second screen open that has nothing but AI on it."* ## [30:05] What Skills Will Actually Matter In The Future? Storytelling tops Galloway's list — the ability to look at data, construct a narrative arc, and communicate it compellingly across every medium. He holds up Jeff Bezos's 1997 shareholder letter, Jensen Huang's stadium keynotes, and Alex Karp's walk-and-talk earnings calls as models. Relationships are the second pillar: as technology converges and products commoditise, the differentiator is whether people want to work with you. He is honest that predicting specific skills is unreliable — private schools doubled down on computer science and Mandarin a decade ago, and neither bet has paid off as expected. > *"The enduring skill is storytelling — your ability to look at data, create a narrative arc and then communicate that story in a compelling way via all the different mediums."* ## [33:45] Are Young People Losing The Ability To Handle Rejection? Galloway identifies the erosion of rejection-tolerance as the most underrated threat facing young people, especially young men. Frictionless online relationships offer a simulacrum of connection without the emotional labour of real-world risk. He mentors young men by assigning deliberate rejection exercises: approach a stranger for friendship, ask someone out for coffee. The goal is not the yes; it is learning that a no is survivable. He argues his own superpower is simply the willingness to mourn failure and try again. > *"The secret to my success is rejection. I ran for sophomore, junior, and senior class president of my high school. I lost all three times."* ## [39:55] Can You Trust The People Building AI? A sharp cultural critique: America has replaced declining religious institutions with tech idolatry, crowning each new CEO as a secular Jesus Christ. Steve Jobs, then Zuckerberg, then Sam Altman, now Dario Amodei — each is briefly positioned as the good guy before completing the villain's journey. Galloway's argument is not that these people are evil but that they are doing exactly what capitalism demands: maximising earnings regardless of wider harm. The answer is not more trustworthy tech founders; it is competent elected officials who regulate them. > *"Can we trust Sam Altman? No. But we shouldn't need to trust him. We should be able to trust that we have smart elected officials that will regulate these companies."* ## [44:50] Are Tech Leaders Quietly Preparing For The End? Galloway reveals that roughly one in three billionaires maintain a "go bag" — a fully funded escape plan, typically a private jet to Auckland and a fortified New Zealand bunker. He calls this nihilism: the ultra-wealthy have sequestered themselves so completely from ordinary infrastructure — private aviation, concierge medicine, private security, elite schools — that they are no longer invested in the health of society. Their disproportionate political donations are therefore not directed at making the system work for everyone. > *"The problem is the 0.1% are not invested in the health of America. They don't have to put up with TSA lines. They fly private."* ## [52:00] Do Some AI Leaders Believe The Risk Is Worth It? A secondhand but chilling account: a source with direct access to an AI CEO described someone who genuinely believes there is a roughly 7–10% chance their work ends in catastrophe, but considers being the person who summoned this new intelligence consequential enough to proceed regardless. Galloway connects this to widening inequality — the delta between middle-class and ultra-wealthy life has expanded so dramatically across healthcare, travel, and security that the incentives of the 0.1% are structurally misaligned with the rest of society. > *"The bottom 99% of Western societies are essentially being optimised and monetised to make the life of the 1% just unbelievable."* ## [58:04] Ads Sponsored segments for LinkedIn Hiring Pro and Function Health. ## [60:05] Could AI Make Us More Human? Galloway offers a surprising positive: unlike social media algorithms that push users toward political extremes, AI models appear to moderate views by seeking statistical medians. He sees genuine value in AI companionship for isolated elderly users. But he returns to his central fear: the biggest downside of AI is not weapons, not election contamination, not even income inequality — it is loneliness. Men aged 20 to 30 are spending less time outdoors than prison inmates, and 42% of men aged 18 to 24 have never asked a woman out in person. > *"The biggest downside of AI in my view is loneliness. AI is convincing people they can have a reasonable facsimile of life on a screen with an algorithm."* ## [65:00] What Happens When AI Becomes Your Closest Companion The conversation shifts to the Iran conflict as a case study in what happens when strategic incompetence meets operational excellence. Galloway credits the initial military strike as tactically credible but argues the absence of Congressional briefing, Gulf ally coordination, and clear exit objectives has produced a quagmire — and notes Iran's IRGC-produced propaganda is outperforming US information operations in the global war of memes. > *"The problem with wars is that the enemy has a say. And all the enemy needs to do — whether it's the Viet Cong or the Taliban or the IRGC — is survive, and they win."* ## [70:00] The Hidden Trade-Off Between Convenience And Real Relationships Galloway diagnoses America's Iran strategy as a product of a gutted diplomatic corps. When senior officials fly to Islamabad expecting a deal, 97% of the preparatory work that career diplomats would normally complete simply has not happened. The IRGC understands the game better: all they need to do is survive, and every day the conflict continues they look like the underdog who stood up to the superpower. His most optimistic scenario is a multinational force enforcing freedom of navigation through the Strait of Hormuz. > *"Do you know what we have done in the US to our diplomatic corps? We've absolutely gutted it."* ## [75:00] Why Loneliness Could Explode US stock markets hit an all-time high during active Middle East conflict — a sign that the wealthy are so insulated from geopolitical risk that war no longer registers in asset prices. The top 10% account for 50% of consumer spending, and that cohort does not care if gasoline hits six dollars a gallon. The pain is outsourced to lower-income households and oil-dependent nations. Galloway frames this dissociation from shared risk as one of the most dangerous structural features of contemporary inequality. > *"We've outsourced the downside of war to less wealthy nations who are very oil dependent, to the Gulf, which is incurring damage here."* ## [79:26] The Real Reason Human Connection Might Become More Valuable Extended discussion of AI market valuations and the historical pattern of infrastructure overbuild. Every great infrastructure boom — railroads, electrification, the internet — ended in a crash, and AI capex now constitutes a significant share of US GDP growth. Galloway argues there is a one-in-three chance AI ends up like jet aviation or vaccines: transformative for humanity but impossible to monetise exclusively for a small group of companies, because open-weight Chinese models could commoditise the entire stack through "AI dumping." > *"AI puts AI out of business. And that is if you look at the convergence of the technologies, all the models are converging."* ## [85:00] What This Means For The Next Generation Galloway argues that a market correction might actually benefit younger generations by making assets affordable again. He flags GLP-1 drugs as his technology pick over AI in terms of real-world human impact. His personal investment philosophy at age 61: aggressive diversification, no single position above 3% of net worth, rotation out of overheated US markets into Europe and Latin America. For young people, the only wealth-building path he trusts is compound interest through low-cost index funds, with money automatically invested before it can be spent. > *"The only answer I have is slowly — find out a way to start saving when you're a teenager, 25 bucks a month, then in your 20s 100, then 500."* ## [90:00] How Power, Politics, And AI Are Becoming Intertwined Drawing on his experience losing 70% of New York Times ad revenue in 60 days during 2008, Galloway warns that younger entrepreneurs have never experienced a true recession. He argues that the political class has systematically bailed out asset-owning baby boomers — COVID relief, corporate bailouts, perpetual market support — while denying younger generations the chance to buy assets at distressed prices. Recessions historically created entry points; that mechanism is now deliberately suppressed. > *"Your generation really doesn't know what a recession looks like. Like, everything stops."* ## [95:00] The Dangerous Gap Between Technology And Regulation Personal finance advice combined with a reflection on the limits of prediction. Galloway's investment rule for young people: put money in yourself first, then in relationships, then in diversified index funds. He is honest that picking winning sectors is largely futile, and that anyone claiming certainty does not know. His own investment in Pokemon cards with his son illustrates that the best investments compound in non-financial ways — relationships and shared experience accrue value that conventional ROI cannot measure. > *"The only answer I have is slowly and it requires some discipline. Save money, diversify, compound interest, invest in relationships early."* ## [100:00] What Happens If Governments Can't Keep Up With AI Asked what a 33-year-old should know that a 61-year-old has learned, Galloway offers three lessons: be humble in success because much of it is luck; forgive yourself in failure because much of it is also circumstance; and invest aggressively in relationships in your 30s, because he spent his prime years professionally focused and nearly ended up isolated. He frames every major disappointment as something people later regret not the thing itself but how upset they allowed themselves to be. > *"Nothing's ever as good or as bad as it seems. Be humble when you're successful. And forgive yourself and realise this will pass."* ## [105:00] The Future Of Work, Power, And Who Really Wins Fatherhood as purpose. Galloway confesses he did not want children and did not fall in love with his sons immediately after birth. What changed his view was discovering that fatherhood is the one investment where a positive financial return is structurally impossible — and that is precisely what makes it purposeful. The same logic applies to any cause large enough to demand more than you can ever get back: veterans, activism, caregiving. He closes with frank advice on partnership, timing, and the liberation of having no choice but to lean into your children's interests. > *"Finding your purpose is finding that thing that you can never get a real positive return on. I will never get a positive return for my children."* ## [110:00] Why The Biggest AI Risks Aren't What You've Been Told The final chapter opens with Galloway's emotional description of his sons' contrasting personalities — one a mirror of himself, one a "different species" he observes with fascination. He discusses his book *Notes on Being a Man*, framing it as letters he hopes his boys will read in 30 years. The closing question — the biggest setback and its lesson — draws the most emotionally raw answer of the episode: his mother's death. He says he has not gotten over it and does not want to, because grief is the receipt for love, and he hopes his sons will one day feel the same about losing him. > *"My mother dying. And you can never tell your parents how much you love them too much. The reverse of love is grief."* ## Entities - **Scott Galloway** (Person): NYU Stern Professor of Marketing, serial entrepreneur, author of *The Four*, *The Algebra of Happiness*, and *Notes on Being a Man*; host of the Prof G Pod and Pivot podcast - **Sam Altman** (Person): CEO of OpenAI; used as the primary case study in the recurring tech-leader idolisation and disillusionment cycle - **Elon Musk** (Person): CEO of Tesla, SpaceX, and xAI; discussed as visionary storyteller whose real products (Starlink, SpaceX) are transformative but whose timelines consistently overshoot - **Dario Amodei** (Person): CEO of Anthropic; cited as the current tech industry "good guy" before the inevitable villain turn - **Jensen Huang** (Person): CEO of Nvidia; held up as a model of storytelling-driven CEO performance via stadium keynotes - **OpenAI** (Organization): Developer of ChatGPT; primary subject of fundraising-hype and overvaluation critique - **Anthropic** (Organization): AI safety company; referenced as beneficiary of the "latest hero" investor narrative - **SpaceX** (Organization): Musk's rocket company; flagged as likely destination for capital migrating away from Tesla at IPO - **Amazon** (Organization): Galloway's top large-cap stock pick for 2026 due to robotics leadership and warehouse automation scale - **Tesla** (Organization): Great car company trading at an unjustifiable multiple that will correct when SpaceX IPOs - **GLP-1 drugs** (Concept): Weight-loss and metabolic medications (Ozempic/Wegovy class) that Galloway argues will create more real-world human impact and shareholder value than AI - **AI dumping** (Concept): Galloway's term for China flooding the US with cheap open-weight AI models to undermine American AI valuations and destabilise the economy - **Go bag / billionaire nihilism** (Concept): The practice among roughly one-in-three billionaires of maintaining funded escape plans as a symptom of disengagement from shared societal wellbeing - **Rejection tolerance** (Concept): Galloway's candidate for the most underrated skill of the AI era — the willingness to hear no, mourn briefly, and try again

Jim Fan, lead of Nvidia's embodied AI research, outlines the transition from language-centric models to World Action Models (WAM) that simulate physical reality. He details a roadmap toward the 'Physical Turing Test' and autonomous factories by 2040, driven by video pre-training and human egocentric data scaling. ## [00:00] Introduction Host Sonya Huang introduces Jim Fan, who leads Nvidia's embodied autonomous research group. Fan reflects on his early days as an intern and the excitement surrounding the future of robotics. > *robots are just one of the most thrilling things that's going to happen.* > *[0, 12]* ## [00:30] DGX One Origin Story Jim Fan recounts the 2016 delivery of the first DGX-1 by Jensen Huang to Elon Musk and the OpenAI team. He highlights how this moment catalyzed the deep learning revolution that led to current AI breakthroughs. > *If you believe in deep learning, deep learning will believe in you.* > *[1, 26]* ## [01:42] The Great Parallel Fan proposes 'The Great Parallel,' applying the successful LLM scaling playbook to robotics. Instead of predicting the next token in a string, the goal is to predict the next physical world state through simulation and alignment. > *instead of simulating strings can we simulate next physical world state?* > *[2, 56]* ## [03:31] Robotics Endgame Setup The strategy for achieving the robotics end game is divided into two primary pillars: model strategy and data strategy. Fan notes that while LLMs are in their final 'boss fight,' robotics is just beginning its scaling journey. > *It boils down to two things, model strategy and data strategy.* > *[3, 32]* ## [03:39] Why VLA Falls Short Visual Language Action (VLA) models are criticized for being 'head-heavy' on language while lacking a fundamental grasp of physics and verbs. Fan argues they are better at encoding static knowledge than dynamic physical interaction. > *VLAs are great at encoding knowledge and nouns, but not so much at physics and verbs.* > *[4, 8]* ## [04:32] Video World Models Fan explains how video models like VEO3 learn internal physics—such as gravity and buoyancy—simply by predicting pixels at scale. These models act as simulators that can solve mazes and plan visual sequences internally. > *Physics emerge by predicting the next blob of pixels at scale.* > *[5, 15]* ## [06:09] DreamZero World Action Nvidia introduces 'Dreamer' and World Action Models (WAM), which jointly decode future world states and motor actions. This allows robots to perform zero-shot tasks by 'dreaming' the correct motion sequence before executing it. > *Dreamer jointly decodes the next world states and next actions.* > *[6, 29]* ## [07:46] Scaling Data Collection To overcome the physical limits of teleoperation, Fan discusses Universal Manipulation Interfaces (UMI) and exoskeletons like Dex-UMI. These tools allow humans to collect high-dexterity data directly without the robot being in the loop. > *we're able to break the curse of 24 hours per robot per day* > *[10, 6]* ## [11:06] EgoScale And Scaling Laws Fan introduces Ego-Exo, a policy trained on 21,000 hours of human egocentric video. This research uncovered a neural scaling law for dexterity, showing a mathematical relationship between pre-training volume and robot performance. > *we discovered this neural scaling law for dexterity.* > *[12, 39]* ## [15:39] DreamDojo And The Roadmap Fan outlines the roadmap to 2040, including the Physical Turing Test and 'lights-out' factories. He introduces Dream Dojo, a neural simulator that replaces classical physics engines with data-driven world models. > *I can say with 95% certainty that we'll get to the end of the end game... by 2040.* > *[19, 19]* ## Entities - **Jim Fan** (person): Lead of the embodied autonomous research group at Nvidia. - **Nvidia** (organization): The technology company developing the hardware and software for the robotics end game. - **Jensen Huang** (person): CEO of Nvidia, mentioned for delivering the first DGX-1 to OpenAI. - **OpenAI** (organization): The research lab that received the first DGX-1 for deep learning development. - **DGX-1** (product): The world's first deep learning supercomputer delivered in 2016. - **VEO3** (model): A video world model capable of simulating physics and visual planning. - **Dreamer** (model): A policy model that predicts future world states and actions simultaneously. - **Ego-Exo** (project): A robotics pre-training framework using large-scale human egocentric video data.

Andrej Karpathy explores the paradigm shift from traditional programming to Software 3.0, where LLMs act as programmable computers driven by context. He details the transition from 'vibe coding' to 'agentic engineering,' emphasizing that while AI handles execution, human taste and understanding remain the ultimate bottlenecks. ## [00:00] Introduction Stephanie Zhan introduces Andrej Karpathy, highlighting his foundational work at OpenAI and Tesla. She notes his unique ability to simplify complex AI shifts and introduces the concept of vibe coding. > *He has a rare gift of making the most complex technical shifts feel both accessible and inevitable. [00:22]* ## [00:44] Feeling Behind as a Coder Karpathy describes a turning point in December 2023 when agentic tools began producing perfect code without manual intervention. This shift led him to adopt vibe coding, trusting the AI to handle complex workflows autonomously. > *I just start to notice that with the latest models the chunks just came out fine. [01:29]* ## [02:28] Software 3.0 Explained Karpathy defines Software 3.0 as a paradigm where the LLM acts as a programmable computer and the context window serves as the primary programming lever. This follows Software 1.0's manual rules and Software 2.0's data-driven weight training. > *Software 3.0 is kind of about your programming now turns to prompting and what's in the context window is your lever. [03:20]* ## [03:44] Agents as the Installer Using the installation of OpenClaw as an example, Karpathy explains how agents replace rigid bash scripts with intelligent, environment-aware execution. This approach allows the AI to debug and adapt to specific system requirements autonomously. > *The agent has its own intelligence that it packages up and then it kind of like follows the instructions. [04:29]* ## [04:49] Menu Gen vs Raw Prompts Karpathy contrasts his custom-coded MenuGen app with raw prompts to models like Gemini, concluding that many traditional software layers are now redundant. He emphasizes that AI can now perform general information processing that was previously impossible with structured code. > *The software 3.0 paradigm is a lot more kind of raw. It just your neural network is doing more and more of the work. [06:11]* ## [07:37] What’s Obvious by 2026 Looking toward 2026, Karpathy envisions neural computers that process raw video and audio directly. These systems would use diffusion models to generate dynamic user interfaces, potentially making traditional UI code obsolete. > *You could imagine completely neural computers... a device that takes raw videos or audio into basically what's a neural net. [08:22]* ## [09:41] Verifiability and Jagged Skills AI models develop 'jagged' capabilities, peaking in verifiable domains like math and code due to reinforcement learning rewards. Karpathy notes the paradox where a model can refactor a massive codebase yet fail simple logic. > *state-of-the-art models today will tell you to walk [to a car wash] because it's so close... This is insane. [11:36]* ## [13:39] Founder Advice and Automation Model performance is heavily dictated by the specific data distributions chosen by frontier labs. Karpathy advises founders to explore the 'circuits' of these models to understand their strengths or use fine-tuning to fill gaps. > *we are slightly at the mercy of whatever the labs are doing, whatever they happen to put into the mix. [12:57]* ## [15:46] From Vibe Coding to Agent Engineering While 'vibe coding' raises the accessibility floor, 'agentic engineering' focuses on maintaining professional quality. This discipline involves coordinating powerful but stochastic agents to accelerate development without sacrificing the engineering bar. > *agentic engineering is about preserving the quality bar of what existed before in professional software. [16:07]* ## [25:17] Agents Everywhere and Learning Karpathy advocates for agent-native infrastructure, expressing frustration with human-centric documentation. He argues that while thinking can be outsourced to AI, human understanding remains a critical bottleneck for directing agents. > *You can outsource your thinking, but you can't outsource your understanding. [28:10]* ## Entities - **Andrej Karpathy** (person): AI researcher and former Director of AI at Tesla and founding member of OpenAI. - **Stephanie Zhan** (person): Partner at Sequoia Capital and host of the discussion. - **Software 3.0** (concept): A paradigm where LLMs act as programmable computers via prompting and context. - **Agentic Engineering** (concept): The professional discipline of coordinating AI agents to maintain software quality. - **MenuGen** (project): An app Karpathy built to OCR and visualize restaurant menus, used as a case study. - **OpenAI** (organization): AI research company co-founded by Karpathy. - **Gemini** (ai-model): Google's LLM used in Karpathy's software comparison. - **Vercel** (organization): A cloud platform used by Karpathy to deploy projects.

Ivanka Trump offre un regard sincère sur sa vie, depuis une enfance unique façonnée par des parents célèbres et une attention médiatique intense, jusqu'à sa carrière marquante dans les affaires et le service public. Elle partage les leçons apprises de sa mère, les défis liés à la construction de la confiance, et comment des expériences déterminantes comme le divorce de ses parents et la tentative d'assassinat contre son père ont forgé sa résilience. Trump aborde également sa philosophie de l'intentionnalité, la force d'être sous-estimée, et son parcours de développement personnel à travers la maternité et la thérapie, jusqu'à son engagement avec Planet Harvest. ## [00:00] Pourquoi la confiance ne vient pas facilement et ce que cela révèle Ivanka Trump a appris très tôt, notamment lors du divorce très médiatisé de ses parents quand elle avait neuf ans, à se méfier des relations hypocrites en raison de l'attention médiatique constante et des paparazzi agressifs. Sa mère lui a enseigné la force d'être sous-estimée et l'importance de filtrer le « bruit » extérieur sous pression. Bien qu'elle ait d'abord développé un puissant mécanisme de défense contre la confiance en autrui, elle a depuis cultivé intentionnellement une approche plus ouverte pour tisser des liens plus profonds, en acceptant les risques inhérents. > *ma mère m'a appris qu'être sous-estimée n'est pas une mauvaise chose. C'est en réalité quelque chose de très puissant [00:22]* > *j'ai vraiment appris à me faire davantage confiance aux autres. [05:48]* ## [03:32] Quand on réalise qu'on est différent, que se passe-t-il ensuite Ivanka Trump a réalisé dès son plus jeune âge que sa vie était atypique en raison de l'attention médiatique et de l'examen public constants, un phénomène qu'elle compare à l'exposition amplifiée des enfants sur les réseaux sociaux aujourd'hui. Elle note que ses parents ont fait des efforts pour la protéger, elle et ses frères et sœurs, de cette attention intense. Elle préfère les conversations approfondies aux interviews fréquentes. > *Je pense qu'il y a toujours eu beaucoup d'attention médiatique et d'examen. On le voit, on en fait l'expérience très tôt. [06:24]* > *tout le monde n'a pas je pense l'expérience que nos enfants ont où partout où ils vont les gens ont un appareil d'enregistrement dans les mains [06:40]* ## [05:44] Comment était vraiment sa mère en privé Ivanka Trump décrit sa mère, Ivana, comme une ancienne skieuse nationale disciplinée qui lui a inculqué la valeur du sport, conduisant Ivanka vers le ballet. Elle se souvient d'un souvenir d'enfance inhabituel : Michael Jackson assistant à sa représentation de Casse-Noisette. Malgré ces expériences extraordinaires, son quotidien était ancré par sa grand-mère maternelle, « Bubby », qui lui offrait un amour inconditionnel et l'exprimait à travers la cuisine. > *ma mère était une skieuse incroyable... elle croyait vraiment en l'importance du sport pour cultiver la discipline [07:07]* > *Ma grand-mère... nous a vraiment élevés... elle m'a enseigné un type d'amour inconditionnel et de tendresse [08:44]* ## [11:47] La différence essentielle qui a façonné celle qu'elle est devenue L'éducation d'Ivanka Trump a été profondément marquée à la fois par sa grand-mère aimante, « Bubby », qui lui prodiguait un amour inconditionnel et des soins quotidiens, et par sa mère, Ivana, qui était un modèle de pionnière. Ivana incarnait la force, l'ambition et la résilience, montrant comment poursuivre des objectifs professionnels tout en étant une mère aimante. Ivanka précise que malgré les carrières prenantes de ses parents, ils étaient présents et lui donnaient le sentiment d'être une priorité, sa grand-mère remplissant le rôle traditionnel de personne de référence. > *Ma mère était une pionnière incroyable... un exemple formidable pour moi de force et de résilience, de glamour, de détermination et d'ambition. [11:57]* > *Je n'ai jamais douté que j'étais sa priorité absolue et qu'il était disponible pour moi. [14:42]* ## [15:43] Ce que le divorce de Donald et Ivana Trump a vraiment signifié pour elle Le divorce très médiatisé de Donald et Ivana Trump, qu'Ivanka a appris par un journal à neuf ans, l'a profondément marquée. Elle se souvient de la peur ressentie face à l'attention médiatique intense et des craintes normales d'un enfant lors d'une séparation parentale. Cette période difficile, qui a généré plus de gros titres que le procès d'O.J. Simpson, a forgé un lien unique entre elle et ses frères et sœurs. Plus tard, après le décès de sa mère, Ivanka a acquis une compréhension plus profonde du caractère complexe d'Ivana, façonné par son éducation en Tchécoslovaquie communiste, regrettant de ne pas lui avoir posé plus de questions de son vivant. > *ce divorce a apparemment généré plus de gros titres que le procès d'OJ Simpson. [20:04]* > *le point positif pour moi et mes frères et sœurs c'est que nous nous sommes vraiment soudés d'une manière différente parce que nous traversions cela ensemble. [23:21]* ## [18:27] La réalité d'être la fille de Trump, ce que les gens comprennent mal Être la fille de Donald Trump signifiait affronter un examen public intense dès le plus jeune âge, en particulier lors du divorce de ses parents, ce qui lui a appris une prudence nécessaire envers la confiance. Elle a depuis appris à « trouver le signal dans le bruit » et à éviter les réseaux sociaux belliqueux, privilégiant la paix intérieure. Ivanka souligne l'authenticité profonde de ses parents, et bien qu'elle aborde la communication avec plus de délicatesse, elle maintient un sens aigu de son identité, guidée par la philosophie stoïcienne, pour vivre authentiquement et résister aux pressions extérieures. > *Si je n'avais pas eu cette leçon, je ne sais pas si je serais aussi forte. Cela m'a appris à ne faire confiance à personne. [18:53]* > *Je ne rends pas les coups parce que je ne... crois pas qu'il faille passer son temps et son énergie à être combatif, à plonger dans cette arène particulière et dans le tourbillon nauséabond des réseaux sociaux. [26:19]* ## [23:36] Comment se construire entourée de pouvoir et de célébrité Entourée de pouvoir et de célébrité, Ivanka Trump a trouvé son identité grâce à un développement personnel intentionnel et à l'expérience transformatrice de la maternité, qui l'a « ouverte » et a approfondi sa capacité d'amour. Elle souligne l'importance cruciale de la conscience de soi pour résister aux pressions extérieures et se définir soi-même, plutôt que de « laisser la foule gagner ». Elle applique cette philosophie à l'éducation de ses enfants en favorisant leur individualité, et attribue à ses propres parents le mérite d'avoir autorisé le désaccord respectueux, lui permettant de rester fidèle à elle-même. > *Si vous ne savez pas qui vous êtes, la foule gagne. [29:55]* > *Ils ont créé un environnement où le désaccord était acceptable. [32:44]* ## [30:57] Pourquoi être sous-estimée est devenu son plus grand atout Ivanka Trump a appris de sa mère que le fait d'être sous-estimée peut être un atout considérable. Au début de sa carrière dans l'immobilier, on la jugeait souvent mal, à la fois comme enfant de parents célèbres et comme jeune femme dans un secteur dominé par les hommes. Elle a exploité cette perception, l'utilisant comme motivation pour travailler plus dur et être ultra-préparée, en tirant finalement profit de ceux qui la sous-estimaient. > *ma mère m'a appris qu'être sous-estimée n'est pas une mauvaise chose. C'est en réalité quelque chose de très puissant [00:22]* > *J'ai canalisé cette peur, ce sentiment, et je l'ai utilisé pour me propulser. [35:06]* ## [32:59] Ce qu'elle recherche vraiment lors d'un recrutement et pourquoi c'est important Lors d'un recrutement, Ivanka Trump privilégie les personnes ayant un sens aigu de leur identité, de l'initiative, un bon jugement et une intelligence pratique, car ces qualités innées sont difficiles à enseigner. Elle insiste sur l'importance de travailler avec des « personnes de qualité » en qui elle a confiance et qu'elle respecte, considérant ces attributs comme fondamentaux pour des relations professionnelles réussies et la dynamique globale de l'équipe. > *C'est très difficile d'enseigner aux gens, vous savez, on peut avoir une personne brillante, mais si elle n'a pas un bon jugement ou si elle n'est pas autonome, c'est très difficile de lui donner ça. [38:15]* > *Je ne veux pas travailler avec des gens que je n'apprécie pas, que je ne considère pas comme de bonnes personnes, parce que je ne veux pas passer mon temps avec quelqu'un en qui je n'ai pas confiance ou que je ne respecte pas. [39:00]* ## [37:49] Pourquoi elle a quitté la mode pour le gouvernement Malgré une offre d'emploi prestigieuse d'Anna Wintour chez Vogue à sa sortie de Wharton, Ivanka Trump a poursuivi sa passion de toujours pour l'immobilier. Elle a ensuite bâti une marque de mode prospère, Ivanka Trump.com, qui a atteint près de 800 millions de dollars de ventes annuelles. Toutefois, elle a pris la décision délibérée de fermer cette entreprise florissante pour se conformer aux règles d'éthique gouvernementale lorsqu'elle a accepté la demande de son père de servir dans son administration. Elle considérait cette opportunité comme un privilège et un devoir indéniables envers son pays, malgré les sacrifices personnels et professionnels considérables. > *Nous réalisions près de 800 millions de dollars de ventes annuelles quand j'ai tout arrêté en entrant au gouvernement. [42:30]* > *Je me sens incroyablement privilégiée qu'il nous ait donné l'opportunité de servir un pays que nous aimons tant. [43:30]* ## [41:06] Ce qui s'est vraiment passé quand Trump a décidé de se présenter La décision de Donald Trump de se présenter à la présidence en 2015 a été annoncée lors d'une réunion familiale à Bedminster, surprenant Ivanka par sa rapidité, malgré ses ambitions politiques de longue date, bien que non exprimées, depuis les années 1980. Elle se souvient d'un moment de panique à 16 ans, craignant qu'il ne se présente, avant d'être rassurée. Son entrée dans la politique présidentielle a été un « ajustement radical » pour la famille, élargissant profondément la vision du monde d'Ivanka au-delà de sa « bulle » new-yorkaise et initiant une « aventure extraordinaire » dans le service public. > *Je me souviens d'un moment où j'ai cru que c'était réel. J'avais 16 ans et j'étais en pension et je l'ai appelé... « Ça va ruiner ma vie. » [51:48]* > *sa campagne m'a ouvert les yeux et j'ai réalisé la bulle dans laquelle je vivais [48:02]* ## [46:23] Trump candidat à la présidence, ce qui a tout changé La décision de Donald Trump de se présenter à la présidence a tout changé fondamentalement pour Ivanka, marquant un « ajustement radical » pour toute la famille. Son entrée non conventionnelle en politique, contournant les parcours traditionnels, était comme « boire de l'eau à la lance à incendie ». La campagne a brisé la « bulle » perçue d'Ivanka à New York, élargissant profondément sa vision du monde et la conduisant à embrasser le privilège de servir son pays. > *C'était comme boire de l'eau à la lance à incendie pour nous tous. [47:08]* > *sa campagne m'a ouvert les yeux et j'ai réalisé la bulle dans laquelle je vivais [48:02]* ## [48:52] Publicités Ce segment présente une publicité pour Shopify, une plateforme de commerce en ligne qui simplifie la création de boutiques, la vente sur les réseaux sociaux et la gestion des opérations avec des outils d'IA. Il fait également la promotion de Pipe Drive, un CRM intelligent utilisé par l'animateur, mettant en avant son tableau de bord visuel pour une visibilité claire du processus de vente. > *Shopify, facilite le démarrage car vous pouvez créer votre boutique, vendre sur les réseaux sociaux, accepter les paiements, utiliser des outils d'IA et tout gérer en un seul endroit. [49:22]* > *Pipe Drive est un CRM intelligent et facile à utiliser... il rend votre processus de vente visible grâce à un seul tableau de bord. [50:17]* ## [51:04] A-t-elle jamais pensé que son père irait vraiment jusqu'au bout Bien que Donald Trump ait envisagé de se présenter à la présidence depuis les années 1980, Ivanka affirme que cette ambition n'a pas été explicitement discutée pendant son enfance. Elle se souvient vivement d'un moment à 16 ans où elle a paniqué, croyant que son père se présentait, avant d'être rassurée que ce n'était pas le cas. Elle note que ses positions sur des sujets comme la politique commerciale sont restées constantes au fil des décennies. > *Je me souviens d'un moment où j'ai cru que c'était réel. J'avais 16 ans et j'étais en pension et je l'ai appelé... « Ça va ruiner ma vie. » [51:48]* > *son point de vue est resté constant dans le temps et reste constant à ce jour exactement sur cette question de politique commerciale [52:35]* ## [54:26] Quitter la Maison-Blanche, un soulagement ou autre chose Quitter la Maison-Blanche n'a pas été un soulagement au sens d'un regret, car Ivanka Trump estime avoir « tout donné sur le terrain » et est fière de ses accomplissements durant ses quatre années de service public. Elle considère cette opportunité de servir comme un « privilège extraordinaire » mais n'a aucun désir de retourner en politique, donnant la priorité à ses enfants et refusant de leur faire payer le prix d'une vie publique prolongée. Elle est satisfaite de ses contributions et estime que son père dispose désormais d'une équipe solide pour le soutenir. > *J'ai tout donné sur le terrain, vous savez ? Je ne regarde pas en arrière en disant... je n'ai pas de regrets. [53:33]* > *Ma première responsabilité est d'être leur mère. [56:49]* ## [58:08] Quelqu'un était-il vraiment préparé à la vie à la Maison-Blanche Ivanka Trump admet que rien ne prépare véritablement un individu à l'expérience intense de la politique de haut niveau et de la vie à la Maison-Blanche. Elle a observé que le pouvoir, tout comme la richesse, tend à amplifier les traits inhérents des personnes. Ses interactions avec des dirigeants mondiaux, des monarques aux élus, les ont démystifiés, révélant qu'au fond, ce ne sont « que des gens » avec des difficultés ordinaires, ce qui a finalement dissipé toute intimidation qu'elle aurait pu ressentir. > *Rien ne vous prépare à cette expérience. [58:26]* > *On réalise qu'au bout du compte, les gens sont des gens. [59:03]* ## [59:44] Ce que la tentative d'assassinat a changé à jamais La tentative d'assassinat contre son père en juillet 2024 a radicalement changé la vie d'Ivanka Trump, intensifiant les préoccupations sécuritaires et nécessitant la protection du Secret Service américain. Témoin de l'événement en temps réel avec ses enfants, sa première réaction a été de les protéger, bien qu'elle ait eu l'intuition que son père s'en sortirait. Cette expérience terrifiante, combinée à d'autres alertes de santé familiales, a renforcé sa conviction en la préciosité de la vie et son engagement à choisir la positivité et à valoriser chaque instant, malgré la corrélation troublante entre service public et violence. > *Ma première réaction a été de les détourner. [62:02]* > *Dans la vie, on n'a le choix que dans sa façon de réagir. Et je choisis de voir l'issue positive. [66:05]* ## [1:07:20] À quoi ressemble la vie après s'être retirée de la politique Après s'être retirée de la politique en 2022, la vie d'Ivanka Trump est désormais centrée sur ses jeunes enfants et sa vie de famille privée, car elle trouvait le « monde obscur » de la politique en contradiction avec sa nature. Elle gère les critiques publiques avec la métaphore de « l'aigle et du corbeau », choisissant de s'élever au-dessus de la négativité plutôt que de s'y confronter. Cette période d'examen public intense, incluant l'expérience de mort imminente de son père, a été un « remède » pour sa croissance personnelle, lui enseignant à rechercher la paix intérieure et l'harmonie dans ce qui est sous son contrôle, et à se concentrer sur la gratitude pour les bienfaits de la vie. > *La politique est un monde assez sombre. Il y a beaucoup d'obscurité, beaucoup de négativité, et c'est vraiment en contradiction avec ce qui me fait du bien en tant qu'être humain. [67:45]* > *La réaction de l'aigle face à cela... n'est pas de se tordre et de se retourner pour faire tomber le corbeau ou de se défendre... C'est simplement de s'envoler plus haut. [69:28]* ## [1:11:04] Publicités Ce chapitre correspond à une courte pause publicitaire au sein du podcast. ## [1:14:24] Comment la thérapie a changé sa façon de voir les choses Ivanka Trump a commencé une thérapie à l'âge adulte, la considérant comme un outil d'« inventaire intérieur », motivée par son « état d'esprit orienté vers la croissance » et un désir de traiter des événements de vie significatifs. Les déclencheurs principaux ont été le second diagnostic de cancer de la thyroïde de son mari Jared, son départ de Washington et le décès inattendu de sa mère. La thérapie l'a aidée à prendre soin d'elle-même et à traiter ses émotions plutôt qu'à les compartimenter, changeant finalement sa perspective sur la compréhension de soi et la façon d'avancer. > *J'ai un état d'esprit très orienté vers la croissance... Je cherche toujours à en apprendre davantage sur moi-même et sur le monde [74:35]* > *Jared a été diagnostiqué d'un cancer de la thyroïde pour la deuxième fois. Et puis ma mère est décédée [75:59]* ## [1:20:28] La perte de sa mère et ce que cela lui a appris Ivanka Trump revient sur la mort soudaine et tragique de sa mère, Ivana Trump, en 2022, soulignant l'impact unique d'une perte parentale inattendue. Elle s'est engagée dans un véritable processus de deuil, affrontant l'inconfort et traitant ses sentiments. En tant que parent, elle cherche désormais à transmettre à ses enfants les qualités positives de sa mère tout en évitant consciemment de leur transmettre ses difficultés, ayant acquis une perspective adulte plus claire sur la vie de sa mère. > *Elle a eu une belle vie quand même. [81:07]* > *J'ai vraiment pris le temps de penser à elle non pas à travers les yeux de l'enfant qui l'idolâtrait pleinement mais à travers les yeux d'une adulte qui la voyait clairement. [83:15]* ## [1:26:28] Les 3 règles qui, selon elle, définissent le succès et le bonheur Ivanka Trump croit que le véritable succès et le bonheur sont définis par trois principes clés, particulièrement pour l'entrepreneuriat, qu'elle partagerait avec sa fille Arabella. Premièrement, il faut véritablement aimer ce que l'on fait, car la passion est essentielle à l'engagement. Deuxièmement, l'authenticité est primordiale ; être soi-même et tracer sa propre voie est crucial, car l'imitation mène à l'échec. Troisièmement, et c'est le plus fondamental, il faut cultiver la confiance en soi avant que le monde ne croie en vous, car c'est le point de départ de toute réalisation. Elle note également que l'équilibre traditionnel « vie professionnelle-vie personnelle » est illusoire, préférant rechercher l'alignement avec ses priorités. > *Je n'ai jamais vu quelqu'un au sommet de son art qui n'aime pas absolument ce qu'il fait. [92:46]* > *vous allez devoir croire en vous avant que le monde ne croie en vous. [94:48]* ## [1:28:37] Ce qu'est Planet Harvest et pourquoi cela pourrait compter plus qu'on ne le pense Planet Harvest est l'entreprise à mission d'Ivanka Trump visant à réduire le gaspillage alimentaire et à soutenir les agriculteurs américains. L'initiative a été inspirée pendant la pandémie de COVID-19 lorsqu'elle a observé d'énormes quantités de produits périssables jetés en raison de problèmes de chaîne d'approvisionnement. Planet Harvest s'attaque au problème persistant d'aliments parfaitement consommables rejetés par les distributeurs pour ne pas répondre à des critères esthétiques stricts, offrant ainsi des revenus supplémentaires aux agriculteurs et bénéficiant à l'environnement. > *Planet Harvest est née... pour s'assurer que quand les gens avaient besoin de nourriture, la nourriture dans les champs ne soit pas gaspillée en étant labourée comme nous l'avons vu au début de la pandémie. [89:18]* > *400 millions de livres de fraises chaque année restent dans les champs... Non pas parce qu'elles sont imparfaites. Elles ne répondent simplement pas à des spécifications esthétiques très rigides. [90:57]* ## Entités - **Ivanka Trump** (Personne) : Fille de Donald et Ivana Trump, femme d'affaires et ancienne responsable gouvernementale. - **The Diary Of A CEO** (Organisation) : Le podcast hébergeant l'interview. - **Donald Trump** (Personne) : Père d'Ivanka Trump, ancien président des États-Unis. - **Ivana Trump** (Personne) : Mère d'Ivanka Trump, ancienne skieuse pour la Tchécoslovaquie. - **Michael Jackson** (Personne) : Célèbre chanteur, auteur-compositeur et danseur américain. - **O.J. Simpson** (Personne) : Ancien joueur de football américain, présentateur, acteur et criminel condamné. - **Marcus Aurelius** (Personne) : Empereur romain et philosophe stoïcien. - **Shopify** (Organisation) : Plateforme de commerce en ligne pour créer des boutiques. - **Pipe Drive** (Organisation) : Logiciel CRM intelligent (Gestion de la Relation Client). - **Anna Wintour** (Personne) : Rédactrice en chef de Vogue. - **Vogue** (Organisation) : Magazine de mode et de style de vie. - **Wharton School of Business** (Organisation) : École de commerce de l'Université de Pennsylvanie. - **Office of Government Ethics** (Organisation) : Agence gouvernementale américaine chargée de prévenir les conflits d'intérêts. - **Jared Kushner** (Personne) : Mari d'Ivanka Trump, ayant également servi au gouvernement. - **US Secret Service** (Organisation) : Agence gouvernementale chargée de la protection d'Ivanka Trump et de sa famille. - **Planet Harvest** (Organisation) : Entreprise cofondée par Ivanka Trump, axée sur la réduction du gaspillage alimentaire et le soutien aux agriculteurs. - **Arabella** (Personne) : Fille aînée d'Ivanka Trump. - **Stoïcisme** (Philosophie) : École de philosophie de la Grèce antique. - **Bouddhisme** (Philosophie) : Philosophie orientale. - **Taoïsme** (Philosophie) : Philosophie orientale. - **Tchécoslovaquie** (Lieu) : Ancien pays d'Europe centrale. - **New York** (Lieu) : Grande ville des États-Unis. - **Bedminster, New Jersey** (Lieu) : Endroit où se trouvait Ivanka Trump lorsqu'elle a appris la tentative d'assassinat contre son père. - **Child Tax Credit** (Politique) : Crédit d'impôt américain pour les familles avec enfants. - **Great American Outdoors Act** (Politique) : Législation soutenue par Ivanka Trump. - **Législation contre la traite des êtres humains** (Politique) : Législation sur laquelle Ivanka Trump a travaillé pendant son service public. - **Formation professionnelle et technique** (Initiative) : Programmes promus par Ivanka Trump pour former et requalifier les travailleurs américains. - **Méditations** (Livre) : Recueil d'écrits personnels de Marc Aurèle.

Présentation en trois minutes par Anthropic de la boucle qu'ils considèrent comme l'habitude la plus importante lors du travail avec Claude Code : d'abord rechercher en mode plan, définir ce que "terminé" signifie avant de toucher un seul fichier, puis faire réviser le diff par un sous-agent avant de pousser. ## [00:03] Pourquoi explorer-planifier-coder-committer surpasse le démarrage immédiat L'introduction est directe : si vous n'adoptez qu'une seule habitude du cours, que ce soit ce workflow. Le mode d'échec combattu est le réflexe de coller une tâche dans Claude et de le regarder générer du code immédiatement, ce qui accélère le démarrage mais repousse le coût de correction. > *Without this, most people jump straight to pasting in Claude to write code, which means more course correcting later on.* ## [00:21] Le mode plan : recherche en lecture seule avant toute édition Le mode plan compresse exploration et planification en un seul geste. Claude peut lire des fichiers et effectuer des recherches web mais ne peut pas écrire — Shift+Tab permet de basculer depuis l'invite. Le narrateur fait la démonstration avec une vraie demande (ajouter la conversion WebP à un pipeline d'upload d'images, identifier l'endroit approprié, les dépendances nécessaires et l'approche). Claude retourne un plan ; vous le lisez et demandez des révisions si quelque chose manque. C'est l'endroit le moins coûteux de tout le cycle pour changer de direction, car rien n'a encore été écrit. > *With plan mode, Claude can't edit files. It just reads files to gather research on how to tackle this implementation.* ## [01:11] Approuver le plan puis corriger la trajectoire pendant que Claude code Une fois le plan satisfaisant, Approuver rend l'exécution à Claude pour qu'il coche les éléments de la liste. Vous choisissez si les éditions de fichiers s'auto-acceptent ou demandent une confirmation à chaque fois. Claude résoudra les problèmes seul, mais prévoyez d'intervenir — et le bénéfice du mode plan ici est que l'agent conserve le contexte de recherche qui a produit le plan, donc les corrections en cours d'exécution atterrissent au bon endroit plutôt que de repartir de zéro. > *This is the benefit of working with plan mode because after the plan is finished, we also have the context of how it got to the results to help it guide its next decision.* ## [01:39] Expliciter les critères de succès et donner de vrais outils à Claude Un plan sans définition du "correct" laisse Claude deviner. Précisez à quoi ressemble le succès, puis équipez l'agent pour le vérifier réellement : l'extension Claude+Chrome lui permet de piloter un onglet de navigateur pour tester une interface qu'il vient de construire ; une suite de tests lui fournit quelque chose à valider à chaque boucle, et Claude peut aussi écrire les tests — mais seulement si vous les avez déjà validés comme vérité terrain. Conseil de durabilité : quand Claude bute sans cesse sur le même problème, demandez-lui de persister la solution dans le fichier CLAUDE.md pour arrêter de le réapprendre. > *In order for Claude to be confident in its results, it has to be clear on what it deems correct.* ## [02:24] Revue par sous-agent, commit et récapitulatif Avant de pousser, lancez un reviewer de code sous-agent sur le diff — une seconde passe sans attachement à l'implémentation. Ensuite, demandez à Claude de rédiger le message de commit dans votre style et d'expédier. Le récapitulatif recadre chaque étape : Explorer fournit le contexte, Planifier définit le succès, Coder est l'aller-retour qui converge vers le plan, Committer révise et pousse pour passer à la suite. > *A tip before you commit, run a sub agent code reviewer to look at your code.* ## Entities - **Anthropic Tutorial Narrator** (Person): La voix officielle d'Anthropic pour le cours Claude Code 101. - **Claude Code** (Software): Outil de codage agentique en terminal dont la boucle quotidienne recommandée est le sujet de cet épisode. - **Plan mode** (Feature): Mode lecture seule activé par Shift+Tab — Claude recherche et propose un plan mais ne peut pas éditer de fichiers. - **Claude + Chrome extension** (Software): Permet à Claude Code de piloter un onglet Chrome pour vérifier les modifications d'interface avant de déclarer une tâche terminée. - **CLAUDE.md** (File): Fichier de mémoire du projet utilisé ici comme cible de persistance pour les corrections que Claude réapprend sans cesse. - **Subagent code reviewer** (Pattern): Sous-agent Claude pré-commit qui révise le diff avant que l'humain pousse.

Le tutoriel Claude Code 101 d'Anthropic sur le contexte — ce qui remplit la fenêtre, quand la compaction automatique se déclenche, et les leviers pratiques (/compact, /clear, /context, claude.md, toggles MCP, skills, sous-agents) pour garder une session suffisamment légère. ## [00:03] Pourquoi le contexte est fini et pourquoi cela compte Le contexte est la mémoire de travail de Claude : chaque prompt, chaque lecture de fichier, chaque résultat d'appel d'outil atterrit dans la même fenêtre. La fenêtre est grande mais finie, donc optimiser ce qui y entre devient incontournable dès que l'on commence des sessions multi-étapes. > *Every file it reads, every command it runs, every message you send, it all takes up space in the context window.* ## [00:39] La compaction automatique et la commande /compact En approchant de la limite, Claude Code compacte automatiquement : il résume les éléments importants et supprime les résultats d'appels d'outils bruyants pour libérer de l'espace. Vous pouvez aussi déclencher `/compact` manuellement — utile quand vous voulez de la marge tout en conservant le fil de votre travail. Compromis : la compaction peut faire perdre les détails des tours précédents. > *Compaction will summarize important details and remove the unnecessary tool call results and free up a lot of space in your context window.* ## [01:11] /clear et /context : repartir de zéro et voir ce qui est utilisé Si vous voulez une réinitialisation totale sans mémoire de la session précédente, `/clear` efface tout. Pour voir où va réellement votre espace, `/context` affiche la taille totale, les catégories les plus gourmandes et un graphique de répartition — le diagnostic avant de choisir entre compact et clear. > *To check the state of your context, run the /context command.* ## [01:35] La règle empirique : compact en cours de tâche, clear entre les tâches Le narrateur donne une heuristique claire : vous travaillez encore sur une fonctionnalité et approchez du plafond ? Compactez — vous voulez que l'historique pertinent persiste. Vous avez terminé le plan et passez à autre chose ? Effacez — l'ancienne conversation risque de biaiser le nouveau travail. > *If you have finished the plan and want to start on a new feature, then clear. You don't want the previous conversation to present bias in anything new that you want to create.* ## [01:57] claude.md, précision des prompts et écrire moins en écrivant plus Tout ce que Claude doit retenir d'une session à l'autre appartient à `claude.md` pour éviter qu'il redécouvre les mêmes faits à chaque fois. Et paradoxalement, les prompts courts coûtent plus de contexte : face à une demande vague, Claude parcourt le codebase avec grep et raisonne davantage, ce qui remplit la fenêtre. Une ou deux phrases de précision supplémentaires économisent beaucoup d'espace par la suite. > *The irony behind writing a smaller prompt is that it in the long run, it will take up more context.* ## [02:26] Serveurs MCP, skills et sous-agents comme outils de gestion du contexte Les serveurs MCP chargent par défaut tous les outils qu'ils exposent dans le contexte — pratique s'ils sont pertinents, coûteux sinon, donc désactivez ceux sans rapport avec le projet. Les skills se comportent comme des serveurs MCP mais n'injectent pas toute leur surface dans le contexte. Les sous-agents fonctionnent en parallèle avec leur propre fenêtre séparée ; pour les tâches de recherche d'informations ("où sont les endpoints d'authentification ?"), vous pouvez déléguer à un sous-agent et ne récupérer que la réponse, sans tout le cheminement. > *Sub agents run in parallel with your main agent but has a complete separate context window.* ## [03:06] Récapitulatif Gérer le contexte dans Claude Code est la différence entre une session longue et productive et une session bloquée. Utilisez `/compact` pour résumer les longues sessions, `/clear` pour repartir à zéro, soyez précis dans vos prompts, vérifiez `/context` pour voir ce qui consomme la fenêtre, et déléguez les tâches de recherche d'information aux sous-agents. > *Managing context within cloud code is crucial. Use slash compact to summarize long sessions and slashclear to start fresh.* ## Entités - **Anthropic Tutorial Narrator** (Person): La voix officielle d'Anthropic pour la série de tutoriels Claude Code 101. - **Claude Code** (Software): L'assistant de codage agentic en terminal d'Anthropic dont la fenêtre de contexte est le sujet de cet épisode. - **Context window** (Concept): La mémoire de travail de Claude — finie, remplie par les prompts, lectures de fichiers et résultats d'appels d'outils. - **/compact** (Command): Commande slash (et déclencheur automatique) qui résume l'historique et supprime le bruit des appels d'outils pour libérer de l'espace. - **/clear** (Command): Commande slash qui efface intégralement la session pour un départ propre sur un nouveau travail. - **/context** (Command): Commande slash qui indique la taille totale du contexte et les catégories qui le consomment. - **claude.md** (File): Fichier mémoire au niveau du projet que Claude lit entre les sessions pour ne pas redécouvrir les mêmes faits. - **MCP servers** (Software): Fournisseurs d'outils qui chargent par défaut tous les outils exposés dans le contexte — à désactiver quand ils sont sans rapport. - **Skills** (Feature): Alternative allégée aux serveurs MCP qui évite de charger toute la surface d'outils dans le contexte. - **Sub agents** (Feature): Agents parallèles avec leur propre fenêtre de contexte, utilisés pour répondre à des questions ciblées sans polluer la fenêtre principale.

Terence Tao évoque le rôle changeant de l'IA en mathématiques et soutient qu'elle automatisera de nombreuses tâches routinières sans remplacer complètement les mathématiciens humains : elle déplacera plutôt leur attention vers de nouvelles frontières. Il insiste sur l'avenir de la collaboration humain-IA et sur la nature imprévisible de l'impact à long terme de l'IA sur la découverte scientifique. ## [00:10] Le rôle actuel de l'IA dans les mathématiques de pointe Terence Tao explique que l'IA effectue déjà des « mathématiques de pointe » que les humains ne peuvent pas faire, même si c'est un autre type de pointe. Il compare cela à la façon dont les calculatrices ont, par le passé, élargi le champ des mathématiques — en prenant en charge, sur un mode spécialisé, des tâches hors de portée humaine. > *D'une certaine manière, elles font déjà des mathématiques de pointe super-intelligentes que les humains ne peuvent pas faire, mais c'est une frontière différente de celle à laquelle nous sommes habitués.* ## [00:52] L'IA comme outil d'automatisation, pas comme substitut Tao prédit que, d'ici une décennie, l'IA gèrera de nombreuses tâches routinières aujourd'hui assurées par les mathématiciens, permettant aux humains de se concentrer sur des problèmes plus complexes et plus importants. Il trace un parallèle avec les bouleversements historiques : les ordinateurs ont automatisé des tâches autrefois confiées à des « calculateurs humains », et le séquençage du génome est devenu automatique sans que la génétique cesse d'évoluer à de nouvelles échelles. > *D'ici une décennie, beaucoup de choses que les mathématiciens font actuellement… pourront être faites par l'IA. Mais nous découvrirons que ce n'était pas la partie la plus importante de ce que nous faisons.* ## [02:46] L'avenir de la collaboration humain-IA en mathématiques Dwarkesh Patel interroge Tao sur la capacité de l'IA à résoudre seule les Problèmes du Prix du Millénaire. Terence Tao estime que « l'hybride humain + IA » dominera les mathématiques bien plus longtemps, car l'IA actuelle ne possède pas encore tous les ingrédients pour remplacer totalement les tâches intellectuelles : elle fonctionne davantage comme un outil complémentaire. > *Je crois vraiment que cet hybride humain + IA dominera les mathématiques pendant beaucoup plus longtemps.* ## [03:43] Un impact imprévisible sur la découverte scientifique Tao reconnaît que, même si l'IA accélérera la science et les découvertes, il est aussi possible qu'elle freine certains types de progrès en « détruisant la sérendipité ». Il conclut que l'impact futur de l'IA sur la découverte scientifique est hautement imprévisible. > *Il est possible que, en détruisant d'une manière ou d'une autre la sérendipité, nous finissions par inhiber certains types de progrès.* ## Entités - **Terence Tao** (Personne) : invité, mathématicien de premier plan de notre époque. - **Dwarkesh Patel** (Personne) : animateur du podcast. - **IA (AI)** (Concept) : intelligence artificielle, abordée dans son rôle en mathématiques et dans la découverte scientifique. - **Mathematica / Wolfram Alpha** (Logiciel) : outils de calcul cités comme exemples d'automatisation en mathématiques. - **Problèmes du Prix du Millénaire (Millennium Prize Problems)** (Concept) : sept problèmes mathématiques non résolus, chacun assorti d'un prix d'un million de dollars.

Les sous-agents sont puissants quand le travail intermédiaire n'a pas sa place dans le thread principal — mais déléguer sans discernement aggrave les choses. Ce tutoriel trace la ligne entre délégation utile (recherche, revue de code, prompts système spécialisés) et anti-patterns courants (revendications d'expertise, pipelines séquentiels, test runners) qui gaspillent le contexte et font perdre les informations dont on a réellement besoin. ## [00:03] Introduction : quand les sous-agents aident ou nuisent La série a jusqu'ici couvert la création et la conception des sous-agents. Ce dernier épisode se concentre sur la question du déploiement : quelles tâches bénéficient vraiment de l'instanciation d'un agent distinct, et lesquelles en souffrent ? La réponse tient à un seul test : le travail intermédiaire compte-t-il pour le thread principal ? Quand l'exploration est séparée de l'exécution, les sous-agents sont rentables. Quand chaque étape dépend de ce que la précédente a découvert, le coût du transfert fait perdre précisément les détails nécessaires. > *"Simply put, the difference comes down to whether the intermediate work matters to your main thread."* ## [00:32] Tâches de recherche : garder l'exploration isolée La traçabilité de l'authentification est un exemple concret. Le thread principal a besoin de savoir où se produit la validation JWT, pas de connaître la douzaine de fichiers parcourus en chemin. Un sous-agent de recherche peut scanner l'ensemble du dépôt, suivre les appels de fonction entre fichiers et renvoyer une réponse unique et précise : la validation JWT se trouve dans middleware/auth.js à la ligne 42, appelée depuis route/api.js. Toute cette exploration reste enfermée dans le contexte du sous-agent. Le thread principal reçoit la conclusion et avance sans que l'historique de recherche n'encombre sa fenêtre. > *"Your main thread receives JWT validation happens in middleware/auth.js at line 42, called from the Express router and route/api.js, or something like that."* ## [01:15] Sous-agents de revue de code : un regard neuf Claude relire du code qu'il a contribué à écrire pose un problème de biais — il a assisté à chaque décision et ne peut pas facilement repérer ce qui paraît suspect de l'extérieur. Un sous-agent reviewer contourne entièrement cela : il ne voit que le diff et les fichiers modifiés, sans aucun historique sur la façon dont le code a évolué. Cette ardoise vierge crée un second avantage. Les critères de revue propres au projet — conventions de nommage, patterns de sécurité, règles architecturales — peuvent être encodés une fois dans le prompt système du sous-agent et appliqués de manière cohérente, sans que le thread principal ait à s'en souvenir tour après tour. > *"A reviewer sub agent sees the changes in a separate context. It runs get diff, reads the modified files, and applies its specialized review criteria without the history of how the code was written."* ## [01:59] Prompts système personnalisés : rédaction et style Le prompt par défaut de Claude Code est optimisé pour une sortie concise et technique — exactement ce qu'il ne faut pas pour une landing page ou un email marketing. Un sous-agent de rédaction reçoit des instructions complètement différentes sur le ton, le public et la structure, produisant un résultat que les réglages par défaut du thread principal ne génèreraient jamais. La même logique s'applique au CSS. Un sous-agent de style qui mentionne les fichiers de votre design system charge automatiquement variables de couleur, conventions d'espacement et patterns de composants dans son contexte avant d'écrire la moindre ligne, garantissant que chaque décision de style reflète le système réel plutôt que des suppositions raisonnables. > *"Claude Code's default prompt tends towards concise, technical writing, which really isn't what you want for a landing page or email campaign, unless you want to put your customers to sleep."* ## [02:57] Anti-patterns : expert claims, pipelines, test runners Trois patterns dégradent systématiquement les résultats. D'abord, les prompts de persona — « Tu es un expert Python » ou « Tu es un spécialiste Kubernetes » — n'apportent rien, car Claude dispose déjà de ces connaissances. Lancer un sous-agent simplement pour lui coller une étiquette d'expert gaspille le surcoût de l'isolation sans rien apporter que le thread principal ne puisse faire. Ensuite, les pipelines séquentiels s'effondrent dès que les étapes ne sont pas vraiment indépendantes. Un flux en trois agents — reproduire un bug, le déboguer, le corriger — paraît propre mais échoue en pratique : l'agent de débogage a besoin du contexte en direct de l'agent de reproduction, pas d'un résumé compressé. Enfin, les sous-agents test runner masquent activement de l'information. Quand les tests échouent, il faut la sortie brute pour diagnostiquer ce qui s'est passé. Un sous-agent qui renvoie seulement « test failed » oblige à écrire des scripts de débogage supplémentaires pour récupérer des détails que la sortie directe aurait montrés immédiatement. > *"A sub-agent that returns a test failed forces you to create additional debug scripts to get details that would have been visible in direct output."* ## [04:10] Récapitulatif et heuristique de décision clé En synthèse : les sous-agents sont des threads isolés créés avec /agents, conçus avec des sorties structurées et des descriptions précises. À utiliser pour la recherche, la revue de code et les tâches nécessitant un prompt système personnalisé. À éviter pour les revendications d'expertise, les pipelines multi-étapes dépendants et l'exécution de tests. Tout le framework se résume à une question : le travail intermédiaire compte-t-il ? Si la réponse est non, déléguez-le. > *"The key question, does the intermediate work matter? If not, then delegate it."* ## Entités - **Anthropic Tutorial Narrator** (Personne) : présentateur de la série de tutoriels Claude Code sur les sous-agents, Anthropic - **Claude Code** (Logiciel) : assistant de codage IA d'Anthropic ; l'environnement dans lequel les sous-agents sont créés et orchestrés - **Subagent** (Concept) : thread Claude isolé lancé depuis le contexte principal, renvoyant un résumé compressé plutôt qu'exposant son contexte de travail complet - **JWT (JSON Web Token)** (Concept) : utilisé comme exemple concret pour un sous-agent de recherche traçant la logique d'authentification dans un dépôt - **System prompt** (Concept) : jeu d'instructions par sous-agent permettant un comportement spécialisé différent du prompt par défaut de Claude Code - **Anthropic** (Organisation) : développeur de Claude et de la série de tutoriels Claude Code sur les sous-agents

Claude Code embarque des sous-agents natifs, mais les sous-agents personnalisés permettent de connecter un comportement spécialisé à des tâches précises. Ce tutoriel crée un sous-agent de révision de code de zéro — en parcourant la commande `/agents`, le choix des outils, le choix du modèle et les champs de configuration qui contrôlent quand et comment Claude délègue. ## [00:03] Ce que sont les sous-agents personnalisés Claude Code inclut des sous-agents natifs, mais vous pouvez aussi créer les vôtres, spécialisés dans des tâches particulières. Un sous-agent personnalisé est un fichier Markdown avec un front matter YAML : le front matter indique à Claude quand router vers cet agent et quelles capacités il possède, tandis que le corps Markdown constitue le prompt système sous lequel le sous-agent s'exécute. > *"Custom sub aents are markdown files with YAML front matter. These markdown files contain configuration that helps claude understand when to use the sub aent and provides directions to the sub aent itself."* ## [00:28] Créer un sous-agent avec /agents La commande `/agents` ouvre le panneau de gestion des agents. En sélectionnant « Créer un nouvel agent », deux questions sont posées : la portée (projet courant ou partagé sur tous les projets de la machine) et la méthode de génération. La voie recommandée est de laisser Claude générer automatiquement l'agent — dans le tutoriel, le narrateur saisit en langage naturel une demande de sous-agent chargé de réviser la qualité du code et les problèmes de sécurité, et Claude s'occupe du reste. > *"Now, the easiest way to create a sub agent is with the / agents command. Next, you can create a sub agent manually, but we recommend using claw code to automatically generate it for you."* ## [00:56] Configurer outils, modèle et couleur Avant que Claude génère le fichier, vous choisissez les outils auxquels le sous-agent peut accéder. Un agent de révision de code n'a pas strictement besoin des outils d'édition, mais laisser l'exécution activée lui permet d'inspecter plus facilement les modifications en attente. Après les outils, on choisit le modèle : haiku pour la vitesse, opus pour la profondeur d'analyse, sonnet pour un équilibre entre les deux. Dernier choix, une couleur qui apparaît dans l'interface pour repérer le sous-agent en un coup d'œil. > *"Now, given that our sub agent is only responsible for reviewing code, you might decide to disallow tools for editing, but I'll leave an execution to allow the sub agent to more easily identify pending changes."* ## [01:43] Comprendre le fichier de configuration Le fichier généré est enregistré dans le projet à l'emplacement affiché dans la fenêtre de résumé. Quatre champs sont essentiels. `name` est l'identifiant unique — vous pouvez l'invoquer en tapant `@agent-code-quality-reviewer` dans un message. `description` est ce que Claude lit pour décider de déléguer ; elle doit tenir sur une seule ligne (les caractères `\n` échappés sont littéraux). Ajouter « proactively » dans la description incite Claude à recourir plus souvent à l'agent ; des exemples de conversations rendent le routage plus précis. `tools` reflète les accès accordés lors de la génération, mais peut être édité directement dans le fichier. > *"If you want Claude to use the sub agent automatically more often, add in the word proactively to the description."* ## [02:41] Le prompt système et son utilisation par Claude Le champ `model` accepte `haiku`, `sonnet`, `opus` ou `inherit` — `inherit` fait tourner le sous-agent sur le même modèle que la conversation parente. Tout ce qui suit le front matter constitue le prompt système : il guide le sous-agent dans sa tâche et lui indique comment rendre les résultats à l'agent principal. > *"The system prompt will provide guidance to the sub agent, helping it understand how to complete its task and how it should return information back to the main agent."* ## [03:15] Tester votre sous-agent Une fois la configuration enregistrée, effectuez quelques modifications de code et demandez à Claude de les réviser. Si le sous-agent ne se déclenche pas au moment attendu, le champ `description` est le premier endroit à ajuster — des exemples plus précis affinent la compréhension de Claude quant au moment de déléguer. > *"If the sub agent isn't being used when you expect, check your description. Adding more specific examples helps Claude understand when to delegate."* ## Entités - **Anthropic Tutorial Narrator** (Personne) : seul présentateur de cet épisode ; anime la série de tutoriels Claude Code subagents sur la chaîne YouTube officielle d'Anthropic - **Claude Code** (Logiciel) : assistant de codage IA d'Anthropic ; prend en charge les sous-agents natifs et les sous-agents personnalisés créés par l'utilisateur - **Custom subagent** (Concept) : fichier Markdown avec front matter YAML qui configure Claude Code pour déléguer des tâches spécifiques à une instance d'agent spécialisé - **/agents command** (Concept) : point d'entrée UI de Claude Code pour créer et gérer les sous-agents ; portée projet ou globale - **System prompt** (Concept) : corps Markdown du fichier de configuration d'un sous-agent ; fournit au sous-agent ses instructions de tâche et de format de sortie à l'exécution - **Anthropic** (Organisation) : créateur de Claude et de la plateforme Claude Code