PodcastsHear the voice. See the shape of the thought.

Figma developer PM Matt Colyer has been building his own AI agents for two years and is buying more software subscriptions than ever — not fewer. He and Every CEO Dan Shipper work through why the "SaaS apocalypse" narrative gets the economics backward, how AI needs to escape the tyranny of the text box to unlock genuinely creative design work, and why the coming year's challenge isn't generation but review: humans are now the bottleneck in a world where agents can ship faster than anyone can evaluate what they made. ## [00:00] AI will create a billion developers This exchange, taken from later in the interview, opens the episode: Matt argues that the number of developers worldwide — roughly 25–40 million a decade ago — is heading toward a billion. That demographic explosion, not AI replacing software, is what makes the SaaS market a "gold mine." Figma and most established SaaS businesses are, in his view, excited rather than threatened. > *"If you're in that space, like, it means it's a gold mine, right?"* ## [01:03] Introduction Dan Shipper frames the conversation: he recently bought Figma stock after noticing the "SaaS apocalypse" discourse, and he wants to know how a company that pre-dates AI is navigating a world where agents can now operate inside your product. Matt, as the director managing Figma's developer products, is the right person to ask. > *"There are all these people who are like, 'Oh, I don't have to use Figma anymore.' You guys just launched an agent in your product. You also have Figma MCP."* ## [02:15] Why the SaaSpocalypse narrative has it backwards Matt's counter-argument runs on two tracks. First, the democratization of software creation massively expands the addressable market — more software being built means more demand for the tools, infrastructure, and services that support it. Second, vibe-coding your own app sounds liberating until you're dealing with SMTP upgrades at midnight. He built his own email agent two years ago and watched it get rickety; these days he pays someone else to run agents for him rather than maintain the plumbing himself. > *"I'm buying more software these days than I ever did before, because I'm like, 'You know what? That tool seems cool. I'm just going to pay somebody else to run my agent for me.'"* ## [05:27] Matt's email agent origin story The origin was unglamorous: three kids in three schools, relentless PTO emails, and the humiliation of missing spirit day. Matt wired up a Python script to grab his inbox and paste it to an LLM — the whole thing was rickety and sometimes the replies didn't work, but the core loop worked. He then added a memory system and a daily summary pushed to him proactively, which he flags as the real unlock: instead of having to open a tool and ask, it just showed up. Dan mirrors this with his own Codex-based inbox workflow, now four weeks into inbox zero. The two also land on voice as an underrated interface — Matt uses Loom recordings because it feels less weird than talking to a blank screen. > *"The unlock for me was like instead of having to go to a tool and ask for the thing, it was just like it would show up."* ## [13:21] Divergent vs. convergent design thinking Chat-based AI is inherently linear — you iterate on one design thread. Matt's argument is that great design has a diamond shape: first you diverge (generate many directions), then you converge (pick the best). Figma's on-canvas agent is a first attempt to break out of the text-box constraint. On the canvas, an agent can spawn a grid of frames — grayscale, sepia, with different type — and then a separate convergent agent can cluster them and recommend which direction to pursue. Command-line agents can't do this kind of spatial, parallel exploration; that's what the canvas unlocks. > *"Text boxes are super limiting — it's very much like a linear 'well this and then that.' If we get to the canvas, the agents allow you to do divergent thinking."* ## [17:39] Figma's MCP server MCP gives third-party agents (Cursor, Windsurf, Claude Code) a standard interface into Figma. Two flows: code-to-design — fire up a dev server, ask the agent to screenshot a live page and pull it into a Figma canvas — and design-to-code via "Get Design Context," which wraps component properties and design library guidelines into an agent prompt that then creates a branch, writes the code, and posts a screenshot to the PR. Both flows remove the manual copy-paste drudgery that used to live between the design file and the codebase. > *"You pull up your codebase, fire up the MCP server, and ask it, 'Hey, can you go to this page and copy it into Figma canvas?' And it will actually do it. That's a little bit mind-blowing."* ## [19:45] Why design agents need personalization Generic agents produce generic output. For Figma, the difference between an okay agent and one people actually love is whether it understands the design system — the components, the spacing rules, the naming conventions. Without that personalization layer, generated designs aren't usable. Matt draws a parallel to the memory systems in chat agents: in Figma's case, the design library is the memory. He also hints at proactive agent work Figma is cooking internally, framing the core problem as maintaining design values at a pace agents can generate. > *"The thing that really differentiates an okay agent from one that people really love is the personalization aspect. For Figma's version of that, it's the design system."* ## [22:09] Every problem is a context problem Matt describes a Figma product operations team that realized every recurring PM task — onboarding docs, project tracking, team introductions — was a context problem in disguise. They built "PMOS": a local SQLite org chart wired to Asana, Slack, and GitHub, then layered Claude Code skills on top. When a new team member joins, the system walks the org chart, reads the last 30 days of Slack channels, checks the Asana board, and produces an uncannily good onboarding file. Dan points out that Claude Code's power comes from the same insight: instead of an always-on cloud agent you have to manually wire to everything, it's an agent that already has access to everything on the user's machine. > *"One of the unlocks to me about AI is like you kind of realize every problem becomes a context problem. The work becomes about framing the problem with the right set of information."* ## [25:12] Apple and Google as the reigning kings of context Matt has been waiting for Apple Intelligence to deliver on its WWDC promise — phones hold all the personal data; an always-on, actually-smart Siri should be the obvious product. It hasn't arrived. He's watching Google's rumored "Spark" agent (always-on, connected to all Google content) with similar anticipation. Dan's take: Apple wins regardless because everyone runs AI on Mac hardware, giving them time to catch up. Matt adds that Apple's privacy-first positioning is a genuine strategic asset, not just PR. > *"Even being late to the game, they are still the king of context. And I think that's what's been interesting to watch about Google I/O this year — seemingly Google has also kind of woken up to that."* ## [28:18] Why review is the new bottleneck Generation is no longer the hard part. Agents are cheap, capable, and available; the problem is that humans are now inundated with net-new content they need to evaluate and approve. Matt frames "review" as the coming year's core design challenge: how do you scale a human value system — what good looks like, what fits your brand — at the pace agents can ship? The format is still unsettled: video walkthroughs, screenshots, a trusted review agent. He closes with a thought on careers: fundamentals still matter (you need to know what long division is even if you use a calculator), and the people who will thrive are the curious ones who ask how something is put together rather than just accepting the output. > *"We have agents that are capable of producing all this stuff, they're available enough, they're cheap enough. We're just being inundated with new content. The bottleneck is now: how do we scale our value system to evaluate it?"* ## Entities - **Matt Colyer** (Person): Director of Product Management for Developers at Figma; has been building personal AI agents for two years; longtime developer tools practitioner. - **Dan Shipper** (Person): Co-founder and CEO of Every; host of the "AI & I" podcast; active AI agent practitioner (inbox zero via Codex). - **Figma** (Organization): Design and prototyping platform; launched an on-canvas agent and an MCP server; central example in the SaaS-in-the-AI-era discussion. - **SaaSpocalypse / SaaS Apocalypse** (Concept): The narrative that AI will make SaaS software obsolete; both guests argue the opposite — AI expands the developer population and demand for SaaS. - **Diamond-shaped design thinking** (Concept): Divergent phase (generate many options) followed by convergent phase (select the best); Colyer argues current chat-based AI only supports linear/convergent work. - **MCP (Model Context Protocol)** (Concept): Standard interface for third-party agents to connect to tools like Figma; enables code-to-design and design-to-code workflows. - **Figma MCP Server** (Software): Figma's implementation of MCP; supports live page screenshot-to-canvas import and "Get Design Context" design-to-code export. - **Claude Code** (Software): Anthropic's coding agent; referenced as an example of an agent with full local file system context; used by Dan Shipper for inbox management. - **Every** (Organization): AI-focused media and software company; Dan Shipper is co-founder/CEO; runs the "AI & I" podcast series. - **Proactive agents** (Concept): Agents that push summaries or actions to users without being asked; Matt identifies the proactive daily email summary as the unlock that made his agent genuinely useful. - **Review bottleneck** (Concept): The emerging constraint in AI-assisted work where generation is fast but human evaluation/approval capacity is the limiting factor.

Dan Shipper, CEO of Every, delivers a day-zero vibe check on Opus 4.8, arguing Anthropic could have called it Opus 5. The model jumps 30 points past Opus 4.7 on Every's Senior Engineer benchmark, edges out GPT-5.5, tops their internal writing tests at 79.6 vs. 73, and is the first model to produce a genuinely good one-shot slide deck. Two catches temper the enthusiasm: performance degrades sharply below "extra high" reasoning, and the Claude desktop app remains cluttered compared to Codex. ## [00:00] What is Every Every is a 30-person applied AI lab for the future of work—part media outlet, part product studio. Dan opens by explaining the subscription (writing, courses, AI-built tools all in one place at every.to) before rolling into the Opus 4.8 assessment. The plug is brief and context-setting: the team has had beta access for a week, and the rest of the video is what they found. > *"Every is the only subscription you need to stay at the edge of AI."* ## [01:07] Anthropic Is Back: The Headline Case for Opus 4.8 Dan had largely abandoned Claude after Opus 4.7—slow, hard to love, and outpaced by Codex and GPT-5.5 in day-to-day use. Even the most loyal Claude users at Every had started routing work elsewhere. Opus 4.8 breaks that pattern: it scores 63 on Every's Senior Engineer benchmark (30 points above Opus 4.7, one point above GPT-5.5), tops their writing tests, and produced the first one-shot slide deck Dan has called genuinely good. Kieran Klaassen, Every's GM, called it "the most human model he's worked with." The one persistent friction is the Claude desktop app itself. Codex is fast, focused, and ships a clean harness; the Claude app still feels like a product built by three separate teams—chat tab, code tab, co-work tab, each with its own feel. Dan is now splitting time between both apps, which he was not doing before. > *"But honestly, they could have called it Opus 5 cuz this is a really great model."* ## [05:02] Reach Test: Paradigm Shift Ratings from the Every Team Every's reach test asks one question: do you actually open this model when work gets hard? Dan rates Opus 4.8 gold/green—paradigm-shift quality, docked one notch because the Claude app harness is only "okayish to pretty good." Kieran, who runs 50 agents a day, gives a straight gold paradigm-shift, one of the rarest grades the team has assigned. Katie Parrot, a senior staff writer and historical Claude fan, lands at green, splitting her work between Opus 4.8 and Codex. > *"It's very rare to give a paradigm shift grade to a model. So I would pay attention to this."* ## [06:32] Benchmarks: Coding and Writing Numbers On coding, Opus 4.8 hits 63 on the Senior Engineer benchmark—the test feeds the model a vibe-coded codebase and asks it to rewrite from first principles, then scores against two human senior engineers who completed the same rewrite (typically scoring in the 80s–90s). GPT-5.5 sits at 62. On Kieran's LFGbench (real-world tasks: SaaS build, e-commerce site, 3D game landscape), the model writes readable code that bridges technical competence and creativity—the "cozy island" 3D scene is notably richer and more vibrant than GPT-5.5's output. On writing, Opus 4.8 scores 79.6 out of 100 on Every's internal benchmark (intro writing, promo emails, mid-piece paragraphs); GPT-5.5 scores 73. The gap is mainly in AI tells: at high and extra-high reasoning settings, Opus 4.8 produces prose that sounds less like a model. It matches a writer's voice from a single paragraph of context better than any other model Dan has tested. > *"Opus 4.8 scores a 79.6 out of 100 on the writing benchmark. GPT 5.5 is 73."* ## [08:57] Emotional Intelligence, Knowledge Work, and the Verdict Dan uses the model for interpersonal and management work—talking through decisions, pressure-testing his own framing. Opus 4.8's thinking traces show it genuinely cycling through permutations before responding, which makes it feel less like a sycophant and more like a useful counterpart. On knowledge work, it's versatile: code and writing coexist cleanly in a single thread, and the slide deck result is the first one-shot deck Dan would actually send to someone. The verdict: if you're a Claude fan, this model delivers. If Codex converted you, add Opus 4.8 as a parallel tool for writing and knowledge work—it's worth the context switch. The harness gap is real, but the model itself is a banger. > *"If you've been converted to Codex, I highly recommend you at least add it as part of your arsenal."* ## Entities - **Dan Shipper** (Person): Co-founder and CEO of Every; presenter and primary evaluator of Opus 4.8. - **Kieran Klaassen** (Person): GM of Kora at Every; gave Opus 4.8 a straight gold paradigm-shift rating on the reach test. - **Katie Parrot** (Person): Senior staff writer at Every; rated Opus 4.8 green, split between it and Codex. - **Every** (Organization): Applied AI lab and media subscription company focused on AI for the future of work. - **Anthropic** (Organization): Developer of Claude and Opus 4.8. - **Opus 4.8** (Software): Anthropic's latest Claude model; subject of the vibe check. - **GPT-5.5** (Software): OpenAI model used as the primary performance comparison across all benchmarks. - **Codex** (Software): OpenAI coding agent; praised for its clean desktop harness and used as the daily-driver counterpoint to Claude. - **Senior Engineer Benchmark** (Concept): Every's proprietary coding benchmark—rewrites a vibe-coded codebase from first principles and scores against human engineers. - **LFGbench** (Concept): Kieran Klaassen's real-world coding benchmark covering SaaS, e-commerce, and 3D scene generation tasks.

Every unter Dan Shippers Führung ist seit GPT-3 von vier auf dreißig Mitarbeiter gewachsen, betreibt Agenten in nahezu jedem Workflow — und stellt weiterhin ein. In einer Rollenumkehr beim *AI & I*-Podcast befragt COO Brandon Gell den Gründer Dan über dessen 8.000-Wörter-Essay „After Automation". Die zentrale These: Steigende KI-Fähigkeiten erzeugen mehr Bedarf an menschlichem Urteilsvermögen, nicht weniger. Der Mechanismus dahinter: KI macht gestern noch wertvolle Expertise billig und allgegenwärtig, überschwemmt so jedes Feld mit Ergebnissen, die nah dran, aber nicht ganz richtig sind — und genau diese Lücke schafft mehr Arbeit für die Menschen, die sie schließen können. ## [00:00] KI erledigt es — und fragt dann: Was kommt als Nächstes? Dieser Austausch aus dem späteren Verlauf des Interviews bringt die zentrale Spannung der Folge auf den Punkt. Brandon beschreibt den typischen KI-Moment: Man gibt einen Prompt ein, ist beeindruckt, fühlt sich überflüssig — und dann stockt die KI und fragt: „Was soll ich als Nächstes tun?" Dan kontert mit dem Satz, der das gesamte Argument trägt: „Je weiter sich ein Agent von einem Menschen entfernt, desto weniger wertvoll ist er." Beide Ausschnitte stammen aus dem Hauptgespräch (rund 00:11 und 00:35), werden hier vorangestellt, um den Rahmen zu setzen. > *„Je weiter sich ein Agent von einem Menschen entfernt, desto weniger wertvoll ist er."* ## [00:51] Einleitung Brandon erklärt die Rollenumkehr: Er befragt Dan, nicht umgekehrt, und wird dessen These hinterfragen. Dan schildert die Entstehung des Essays: Mitten in einem der agenten-nativsten Unternehmen der Welt beobachtet er, wie Headcount und Automatisierung gemeinsam wachsen — ein Widerspruch zur gängigen Erzählung, KI vernichte Jobs. Der Tweet des ClickUp-CEOs, der einen Großteil seiner Belegschaft entließ und dies mit KI begründete, landet früh im Gespräch als erster Belastungstest: Gilt „After Automation" auch für ein 10.000-Personen-Unternehmen, nicht nur für einen Early-Adopter wie Every? > *„Wenn du in unserem Slack mit einem Stock herumschlägst, triffst du genauso wahrscheinlich einen Menschen wie einen Agenten."* ## [05:51] Das KI-Paradox: mehr Automatisierung, mehr menschliche Arbeit Dan entfaltet sein Kernargument. KI ist auf allen bisherigen Outputs trainiert und kann „gestern noch wertvolle Expertise" günstig an alle liefern. Das demokratisiert Output — Ops-Leute mergen Pull Requests, Nicht-Ingenieure shippen Features — aber das Ergebnis ist gleichförmig nah dran, nie ganz richtig. Es passt nicht zur konkreten Situation. Damit entsteht eine Flut beinahe richtiger Arbeit, die für sich allein an Wert verliert, während gleichzeitig mehr Experten gefragt sind, die diese Arbeit über die Ziellinie bringen. Brandon ergänzt das Every-interne Bild: PRs, die plausibel wirken, bis ein Senior Engineer unter die Haube schaut. > *„Man überschwemmt das Feld mit jeder Menge Zeug, das nah dran ist — aber eben nicht ganz richtig."* ## [10:00] Wie KI gestern noch wertvolle Expertise zur Massenware macht Dan erweitert das Argument um den Benchmark-Einwand: Ja, Modelle verbessern sich exponentiell, aber sobald ein Benchmark gesättigt ist, lässt er sich durch leichte Umformulierung des Problems neu aufspannen. Das tiefere Problem: Menschen besitzen eine Schicht impliziten, nicht artikulierbaren Könnens, die sauberer Spezifikation entgeht — und alles, was sich artikulieren lässt, kann ein Modell optimieren. Everys Erfahrung bestätigt das: Kieran baute in ein bis zwei Monaten ein vollständiges Inbox-Feature von Grund auf, was vorher „völlig unmöglich" gewesen wäre. Der Wert entstand aber daraus, dass ein Experte wusste, was zu bauen war, und jeden Schritt steuerte. > *„Es gibt tatsächlich vieles, was du tust, das sich nicht in einem sauberen Rahmen artikulieren lässt."* ## [18:00] KI handelt autonom, aber sie hat keine Handlungsmacht Brandon zieht die Linie zwischen Autonomie und Handlungsmacht: KI-Agenten werden sehr gut darin, offene Aufgaben ohne Anleitung auszuführen — das ist aber kategorial verschieden von echter Handlungsmacht, dem selbstmotivierten, spielerischen „Ich will das einfach tun, weil es mich interessiert", das selbst ein Kleinkind besitzt. Dan stimmt zu, dass es keinen wirtschaftlichen Anreiz gibt, das zu bauen: Wenn ein Agent am Schreibtisch sagt „Ich hab keine Lust", ist das ein Produktfehler. Der gesamte Anreizapparat der Branche treibt in Richtung Gehorsam und Korrigierbarkeit — genau das, was Menschen im Loop hält. > *„Agent bedeutet, im Auftrag von jemandem zu handeln. Das ist völlig verschieden davon, Handlungsmacht zu besitzen — wie sie selbst das kleinste Kind hat."* ## [20:39] Warum Dan voll auf AGI setzt Brandon schlägt einen Ein-Wort-Antwort-Test vor: Glaubst du, AGI wird kommen? Dan: Ja. Ist das gut? Dan: Ja. Seine AGI-Definition — jeder Agent, den man wirtschaftlich sinnvoll dauerhaft laufen lässt, der kontinuierlich Tokens generiert und Aufgaben erledigt, ohne neu angeprompted zu werden — ist präzise genug, um testbar zu sein. Seine Begründung: Selbst ein wirklich autonomes System wäre gebaut worden, um menschliche Ziele zu erfüllen; anderenfalls würden wir es nicht bauen. Brandons Sorge: Sobald Daueragenten wirtschaftlich rational werden, wird das Massenentlassungsargument kohärent. > *„Jeder Agent, den man nie abschaltet — bei dem es wirtschaftlich Sinn ergibt, ihn ständig laufen zu lassen, der aktiv Aufgaben erledigt, ohne dass man ihn je neu anprompted."* ## [21:57] KI-bedingte Entlassungen sind eine Lüge Dan und Brandon sezieren den ClickUp-Fall: ein CEO, der öffentlich einen Großteil seiner Belegschaft entließ und KI dafür verantwortlich machte. Dans Lesart: Generische SaaS-Unternehmen entlassen Mitarbeiter, wenn sie in Schwierigkeiten sind oder zu aufgebläht wurden — und nennen dann KI als Deckmantel. Brandon ergänzt Jensen Huangs Gegenstück: „Wenn deine Antwort auf Fortschritt das Feuern von Menschen ist, bist du kein besonders kreativer CEO" — selbstdienlich, aber wohl zutreffend. Das ehrliche Bild: KI verändert Workflows tiefgreifend, was unternehmensweite Umstrukturierungen erzwingt. Wer das überspringt und einfach kürzt, nimmt den bequemen Weg. Metas angebliches Keylogging von Mitarbeitern zum Sammeln von Trainingsdaten wird kurz als kreativere, wenn auch beunruhigende Alternative erwähnt. > *„Ich wäre wirklich skeptisch gegenüber jedem, der behauptet, KI werde alle Jobs oder alle Wissensarbeit vernichten."* ## [25:42] Wer mit den Modellen mitschwimmt, kommt durch Selbst unter einem AGI-Szenario ist die entscheidende Variable das menschliche Urteil darüber, was zählt — und was zählt, verändert sich ständig, auch weil KI die Welt selbst immer neu formt. Call-Center-Mitarbeiter in Omaha, die Chatbots misstrauen, oder Unternehmen, die Support-Teams entlassen und zwei Monate später still wieder einstellen, zeigen, wie weit reale Adoption hinter dem Hype herhinkt. Adoption braucht eine Generation, um zu landen; irgendwann werden alle Zugang zu diesen Tools haben; die Gewinner sind die, die mit jedem neuen Modell neu lernen. Dan schließt mit seinem prägnantesten Satz: Wer mit den Modellen mitschwimmt, kommt durch. > *„Wenn du einfach mit den Modellen mitschwimmst — wenn neue Modelle kommen, lernst du, sie für das zu nutzen, was du tust, was auch immer das ist — wirst du gut durchkommen."* ## [35:30] Wie man KI als Lektor für Longform-Texte einsetzt Dan beschreibt den konkreten KI-gestützten Prozess hinter „After Automation". Jeden Morgen hat er den aktuellen Stand seines Arguments in Proof eingesprochen, das Log dann Claude gegeben und gefragt: „Was versuche ich eigentlich zu sagen?" Als die Entwürfe über 4.000 Wörter wuchsen, ließ er Codex die jeweils aktuelle Version in einen Podcast umwandeln und hörte sie auf dem Arbeitsweg — Flussprobleme aufspüren, ganz ohne Hände. Der Text wurde vier- oder fünfmal komplett neu begonnen, bevor das Argument stimmte. Sein Fazit: KI hat den Essay nicht geschrieben, aber sie ermöglichte es, die gesamte 8.000-Wörter-Struktur im Arbeitsgedächtnis zu halten, ohne den Faden zu verlieren. > *„Ohne sie hätte ich das nicht schreiben können. Ich ließ Claude mein Log nehmen und fragte: ‚Was versuche ich wirklich zu sagen?' Und es antwortete — und ich dachte: ‚Ja, genau das versuche ich zu sagen.'"* ## Entitäten - **Dan Shipper** (Person): Mitgründer und CEO von Every; regelmäßiger Gastgeber von *AI & I*; hier der Interviewte, der seinen Essay „After Automation" vorstellt - **Brandon Gell** (Person): COO von Every; moderiert diese Folge als Gastgastgeber und befragt Dan in einer Rollenumkehr - **Every** (Organisation): KI-natives Medien- und Softwareunternehmen; seit GPT-3 von 4 auf 30 Mitarbeiter gewachsen bei gleichzeitig starker Automatisierung; veröffentlicht den *AI & I*-Podcast - **After Automation** (Konzept): Dan Shippers 8.000-Wörter-Essay mit der These, dass KI-Automatisierung die Nachfrage nach menschlicher Expertise steigert, indem Felder mit beinahe korrektem Output überschwemmt werden - **Expertise-Lücke** (Konzept): Die These, dass KI „gestern noch wertvolle Expertise" günstig liefert, aber stets leicht daneben — und damit mehr Bedarf an Menschen schafft, die die Lücke zur realen Situation schließen können - **AGI** (Konzept): In dieser Folge definiert als jeder Agent, den man wirtschaftlich sinnvoll dauerhaft ohne Neu-Prompting laufen lässt; Dan ist überzeugt, dass AGI kommen wird und netto positiv ist - **Autonomie vs. Handlungsmacht** (Konzept): Brandons Unterscheidung zwischen KI, die offene Aufgaben ohne Anleitung ausführt (Autonomie), und KI mit selbstmotivierten Antrieben (Handlungsmacht); Letzteres wird nicht gebaut - **Proof** (Software): Schreibwerkzeug, das Dan für tägliche Voice-Monolog-Entwürfe nutzt; Teil des KI-Feedback-Loops während der Essay-Entwicklung - **Codex** (Software): OpenAI-Tool, mit dem Dan Essay-Entwürfe in Audio-Podcast-Format umwandelte, um sie auf dem Arbeitsweg zu hören - **ClickUp** (Organisation): SaaS-Unternehmen, dessen CEO öffentlich einen Großteil der Belegschaft entließ und dies mit KI begründete; dient als Fallstudie für KI-gewaschene Entlassungen

Noah Brier betreibt Claude Code auf einem Mini-PC in seinem Keller, synchronisiert mit seinem Obsidian-Vault über ein Tailscale-VPN, und erledigt echtes Denken, Recherchieren und Client-Code von seinem Handy aus. Das Gespräch behandelt, wie er diesen Stack aufgebaut hat, warum er strenge "Denkmodus"-Guardrails setzt, um zu verhindern, dass das Modell voreilig Artefakte entwirft, und seine übergeordnete Theorie, dass KI erfolgreich ist, indem sie in die organisatorischen Nischen und Ritzen gelangt, anstatt von den Menschen zu verlangen, neue Strukturen zu übernehmen. Dan Shipper und Noah diskutieren auch, was es wirklich bedeutet, KI-Intuition aufzubauen, und warum Noah glaubt, Kinder auf KI vorzubereiten bedeute weniger, Schummeln zu unterbinden, als episteme Skepsis zu lehren. ## [00:00] Noah Briers Claude Code-Setup auf einem Keller-Server Dan Shipper eröffnet die Episode mit der Beschreibung des Setups, das Noah so interessant macht: ein Heimserver im Keller, der Claude Code auf einem Obsidian-Vault betreibt, von überall per Handy erreichbar. Noah hat das so eingerichtet, dass er denken, recherchieren, schreiben und Code deployen kann, ohne am Schreibtisch sitzen zu müssen. > *"Er hat einen Heimserver im Keller aufgesetzt, sein Obsidian-Vault dort gespeichert und Claude Code darauf gestartet, damit er von seinem Handy aus denken, recherchieren, schreiben und sogar Code deployen kann."* ## [00:52] Einführung Dan und Noah unterhalten sich zum ersten Mal seit etwa fünf Jahren. Noahs Hintergrund umfasst Markenstrategie (er gründete Percolate mit), KI-Beratung bei Alephic und die BRXND.AI-Konferenz. Dan gestaltet das Interview um den praktischen Stack, den Noah aufgebaut hat, statt um abstrakte KI-Diskussionen. > *"Ich freue mich, dich dabei zu haben. Es ist wirklich schön, dich zu sprechen. Das ist unser erstes Interview seit wahrscheinlich fünf Jahren."* ## [02:10] Wie man auf dem Handy Tiefarbeit leisten kann Noah stellt klar, dass sein Setup weniger "Vibe Coding" als strukturierte Wissensarbeit ist. Er gab Evernote zugunsten von Obsidian auf, weil Markdown-Dateien und Ordner etwas sind, auf dem Claude Code tatsächlich operieren kann. Sein primärer Claude Code-Anwendungsfall ist die Interaktion mit seinen Notizen, nicht das Generieren von Code, und die Handy-Erweiterung dieses Setups hat seine Arbeitsgewohnheiten grundlegend verändert. > *"Meine absolute Hauptanwendung für Claude Code ist, es als Werkzeug zur Interaktion mit meinen Notizen zu nutzen."* ## [05:30] Warum Noah Grok für den besten Voice-AI hält Noah bevorzugt Groks Voice-Mode gegenüber den Äquivalenten von OpenAI und Gemini: Gemini war nicht intelligent genug, und der alte GPT-4o-Voice war für seine Zwecke unbrauchbar. Er nutzte ihn auf einer fünfstündigen Solo-Fahrt, um einen Artikel über Transformer durchzuarbeiten, über Bluetooth verbunden wie ein persönlicher Recherche-Podcast. Das Gespräch deckt eine gemeinsame Frustration auf: Voice-Modelle machen noch kein gutes Tool-Calling oder Web-Recherche, was ihre Nützlichkeit für ernsthafte intellektuelle Arbeit einschränkt. > *"Ich habe eine Stunde-Session gemacht, und es war mit Abstand die beste Erklärung, die ich je gelesen oder gehört habe."* ## [11:11] Die technischen Details von Noahs Claude Code-Obsidian-Setup Noah führt durch seinen live gezeigten Obsidian-Ordner. Claude Code sitzt im Obsidian-Root-Verzeichnis, damit es das gesamte Notizarchiv erreichen kann. Für einen Vortrag bei BRXND.AI über das WWII Simple Sabotage Field Manual und Bürokratie in großen Organisationen hat er einen Projektordner in Obsidian angelegt, mit Transkripten aus Chats mit ChatGPT, Claude und Grok sowie Artikeln und PDFs. Claudes Aufgabe ist es nicht, den Vortrag zu schreiben, sondern ihm beim Denken zu helfen: relevante Notizen ziehen, tägliche Fortschritte in ein Log zusammenfassen und klärende Fragen stellen. Die Denkmodus-Einschränkungen legt er explizit im CLAUDE.md-Frontmatter des Projekts fest. > *"Ich bin im Denkmodus, noch nicht im Schreibmodus. Ich habe Claude Code explizit gesagt: Hilf mir gerade nicht beim Schreiben."* ## [26:05] Einen Agent in Claude Code als 'Denkpartner' nutzen Noah argumentiert, dass das Wort "generativ" verzerrt hat, wie Menschen KI nutzen: Alle fokussieren sich auf die Fähigkeit, Artefakte zu produzieren, fast niemand spricht darüber, wie bemerkenswert die Lesefähigkeit ist. Er unterhält einen dedizierten Denkpartner-Agent mit expliziten Guardrails: "Erstelle keine Gliederungen, Entwürfe oder Versionen von Vorträgen oder Texten." Der Agent protokolliert Fragen, verfolgt aufkommende Erkenntnisse und führt ein laufendes Protokoll, damit Noah genau dort weitermachen kann, wo er aufgehört hat. Er verfolgt einen Faden von ChatGPT Deep Research über Wild Bill Donovan bis zu einer vorläufigen Idee über die Verbindung zwischen der Parallelität der Transformer-Architektur und der operativen Autonomie der Special Forces. > *"Weil wir es 'generativ' nennen, liegt viel zu viel Fokus auf der Fähigkeit zu schreiben und zu wenig auf der Fähigkeit zu lesen."* ## [30:23] Noahs Thomas-English-Muffin-Theorie der KI Das Kapitel beginnt mit Noahs Bürokratie-These: Große Unternehmen scheitern nicht bei der Software-Adoption, weil sie faul sind, sondern weil neue Software historisch verlangte, dass sich Organisationen um sie herum restrukturieren. KI sei anders. Sie gelangt in die Nischen und Ritzen der bestehenden Arbeitsweise, daher seine Thomas-English-Muffin-Metapher. Dan fügt ein konkretes Beispiel aus Every hinzu: Zwei Produkte auf unterschiedlichen Stacks mussten eine Dateisuche-Lösung teilen, und Claude Code ermöglichte die Wiederverwendung der Logik, ohne ein gemeinsames Framework zu erzwingen. Das Gespräch weitet sich auf Noahs Idee der "Bürokratie als Positional Encoding" aus. > *"Das ist meine Thomas-English-Muffin-Theorie der KI: Sie gelangt in die Nischen und Ritzen."* ## [39:47] Der noch unerschlossene Raum in der KI Noah und Dan argumentieren, dass die meisten Praktiker, auch gut finanzierte, noch auf fragilen Intuitionen darüber basieren, was diese Modelle tatsächlich können. Noahs Icebreaker bei jedem Client-Meeting ist: "Was war dein Aha-Erlebnis mit KI?", weil dieser Moment der Nicht-Determiniertheit, dieselbe Frage zweimal zu stellen und verschiedene Antworten zu bekommen, wirklich neu ist und Zeit braucht, um internalisiert zu werden. Er borgt Destin Sandlins Rückwärtsfahrrad-Experiment, um den Punkt zu machen: Motorische Intuition und konzeptuelle Intuition sind getrennt. Dan entgegnet, dass Sprachmodelle selbst das Vokabular generieren könnten, das uns fehlt, um über probabilistische Systeme nachzudenken. > *"Wir sind es nicht gewohnt, Dinge zu nutzen, bei denen man dieselbe Frage zweimal stellt und verschiedene Antworten bekommt."* ## [48:44] Wie Noah seine Kinder auf KI vorbereitet Noahs Zehnjährige baute eine Secret-Santa-App mit Claude, die ihr zufällig Datenmodellierung beibrachte: Sie erkannte, dass sie "Gruppen" statt "Erwachsene und Kinder" brauchte. Diese Geschichte verankert ein breiteres Argument: Die Aufgabe von Lehrern ist nicht, KI-Nutzung zu verhindern, sondern Schüler davon zu überzeugen, dass es sich lohnt, Grundfertigkeiten zu erlernen. Er pitcht einen NYU-Kurs namens "Code ist Essay" für den Herbst 2026, und er glaubt, die relevante Meta-Fähigkeit sei episteme Skepsis, also misstrauischer gegenüber Informationen zu sein, die die eigene Meinung bestätigen. > *"Ich glaube nicht wirklich, dass dein Job ist, diesen Kindern das Schreiben beizubringen, denn das ist eine lebenslange Aufgabe. Ich glaube, dein Job ist, sie davon zu überzeugen, dass es sich lohnt, schreiben zu lernen."* ## [01:00:06] Wie er sein Claude Code-Setup aufs Handy gebracht hat Noah demonstriert live den vollständigen mobilen Stack: Termius (SSH-Client auf dem iPhone), Tailscale-VPN zur Verbindung mit dem Keller-Mini-PC, Obsidian synchronisiert über privates GitHub, Claude Code im Terminal. Er zeigt, wie er Claude fragt: "Was gibt es Neues in den letzten zwei Tagen?" und eine Zusammenfassung seiner letzten Obsidian-Aktivitäten bekommt. Er hat auch einen kaputten Link auf seiner Konferenz-Website vom Handy gefixt, das Problem bestätigt, Claude einen PR pushen lassen, fertig. Sein aktuelles Basteln erstreckt sich auf Simon Willisons `llm` CLI-Tool und ein Skript, das alle Anhangsdateien in seinem Obsidian-Vault umbenennt und die Link-Tabelle neu aufbaut. > *"Ich ging raus und saß eine Weile draußen, dann gab es ein Projekt, das an einen Kunden geliefert werden musste, und eine kleine Änderung war nötig. Ich sagte Claude Code genau, wo es schauen soll, bestätigte, dass das Problem das war, was ich dachte, und ließ es eine Lösung pushen. Es hat einen PR gepusht, und dann war ich fertig."* ## Personen - **Dan Shipper** (Person): CEO und Mitgründer von Every; Host des Interviews - **Noah Brier** (Person): Mitgründer von Percolate; Gründer der KI-Strategieberatung Alephic; Organisator der BRXND.AI-Konferenz - **Every** (Organisation): Medien- und Softwareunternehmen, das diesen Podcast produziert - **Alephic** (Organisation): Noahs KI-Strategieberatung; arbeitet mit Fortune-50-Kunden darunter Amazon, Meta und PayPal - **BRXND.AI** (Organisation): Jährliche Konferenz an der Schnittstelle von Marketing und KI, organisiert von Noah; 2025 Edition am 18. September in New York City - **Claude Code** (Software): Anthropics agentisches Coding-Tool; zentral für Noahs Second-Brain- und mobilen Workflow - **Obsidian** (Software): Markdown-basierte Notiz-App; Noahs primärer Wissensspeicher, nach der PARA-Methode organisiert - **Tailscale** (Software): Mesh-VPN zur sicheren Verbindung von Noahs Handy mit seinem Keller-Mini-PC - **Termius** (Software): iOS-SSH-Client, den Noah nutzt, um von seinem Handy auf seinen Heimserver zuzugreifen - **Grok** (Software): xAIs KI-Assistent; Noah hält seinen Voice-Mode für deutlich besser als OpenAIs und Geminis für substanzielle Recherche - **Simple Sabotage Field Manual** (Konzept): WWII-zeitliches OSS-Dokument, das Noah neu veröffentlicht hat; als Linse auf moderne organisatorische Bürokratie in seinem BRXND.AI-Vortrag - **Thomas-English-Muffin-Theorie** (Konzept): Noahs Metapher dafür, wie KI erfolgreich ist, indem sie in bestehende organisatorische Workflows passt, anstatt Restrukturierungen zu verlangen

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.

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