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, la empresa de Dan Shipper, creció de cuatro a treinta personas desde GPT-3, integra agentes en prácticamente todos sus flujos de trabajo y sigue contratando. En un giro de formato para el programa *AI & I*, el COO Brandon Gell entrevista a Dan sobre su ensayo de 8.000 palabras "After Automation", que sostiene que el avance de la IA aumenta la demanda de juicio humano, no la reduce. El mecanismo central: la IA abarata y masifica la competencia experta de ayer, inundando cada sector con resultados que están cerca pero no del todo bien — y esa brecha genera más trabajo para quienes saben cerrarla. ## [00:00] La IA lo hace y luego pregunta qué sigue Este intercambio, tomado de más adelante en la conversación, captura la tensión central del episodio. Brandon describe el momento arquetípico con la IA: la interrogas, te deja sin palabras, te sientes obsoleto — y entonces se detiene y pregunta: "¿Qué hago ahora?" Dan responde con la frase que ancla todo el argumento: "Cuanto más lejos está un agente de un humano, menos valioso es." Ambos fragmentos provienen de la conversación principal (alrededor de los minutos 00:11 y 00:35), y se presentan aquí para enmarcar lo que sigue. > *"Cuanto más lejos está un agente de un humano, menos valioso es."* ## [00:51] Introducción Brandon explica el giro de formato: esta vez él entrevista a Dan, no al revés, y va a cuestionar su tesis. Dan cuenta el origen del ensayo: estar dentro de una de las empresas más nativas en el uso de agentes del mundo, ver cómo el equipo crece junto con la automatización y sentir una disonancia con el relato dominante de que la IA está eliminando empleos. El tuit reciente del CEO de ClickUp — que despidió a buena parte de su plantilla achacándoselo a la IA — entra en la conversación como el primer test del argumento de Dan: ¿aplica "After Automation" a una empresa madura de 10.000 personas, no solo a una como Every que adopta todo desde el principio? > *"Si agitas un palo en nuestro Slack, tienes las mismas probabilidades de darle a un humano que a un agente."* ## [05:51] La paradoja de la IA: más automatización, más trabajo humano Dan desarrolla el argumento central. La IA se entrena con todos los resultados anteriores, por lo que puede ofrecer "la competencia experta de ayer" de forma barata y universal. Eso democratiza el output — alguien de operaciones hace merge de pull requests, alguien sin perfil técnico lanza features — pero el resultado es uniformemente *cercano, no correcto*. No está ajustado a la situación concreta. Se genera una avalancha de trabajo casi correcto que se devalúa solo, y al mismo tiempo crece la demanda de expertos capaces de llevar ese trabajo a buen puerto. Brandon aporta la versión interna de Every: PRs que parecen razonables hasta que un ingeniero senior mira bajo el capó. > *"Inundas la zona con montones de cosas que están cerca, pero no del todo bien."* ## [10:00] Cómo la IA abarata la competencia experta de ayer Dan extiende el argumento a la objeción de los benchmarks: sí, los modelos mejoran de forma exponencial, pero en cuanto un benchmark se satura siempre puedes desaturdirlo reformulando el problema. El fondo del asunto es que los humanos tienen una capa de competencia tácita, no articulada, que escapa a cualquier especificación limpia — y todo lo que sí puedes articular, un modelo puede escalarlo. La experiencia de Every lo confirma: Kieran construyó de principio a fin una funcionalidad completa de bandeja de entrada en uno o dos meses, algo "completamente imposible" antes. Pero el valor vino de un experto que sabía qué construir y dirigía cada paso. > *"Hay mucho de lo que haces que no puede articularse en un marco limpio."* ## [18:00] La IA puede actuar de forma autónoma, pero no tiene agencia Brandon traza la línea entre autonomía y agencia: los agentes de IA son cada vez mejores ejecutando tareas abiertas sin supervisión constante, pero eso es cualitativamente distinto de tener *agencia* — esa motivación propia, lúdica, de "quiero hacer esto porque me apasiona" que tiene hasta un niño pequeño. Dan coincide en que no hay incentivo económico para construir eso: si estás en tu mesa y el agente dice "paso, estoy jugando", eso es un fallo de producto. Toda la estructura de incentivos del sector empuja hacia la obediencia y la corregibilidad, que es exactamente lo que mantiene a los humanos en el bucle. > *"Agente significa algo que actúa en nombre de otro. Eso es muy distinto de tener agencia, que es lo que tiene incluso el niño más pequeño."* ## [20:39] Por qué Dan apuesta todo a la AGI Brandon propone el test de una sola palabra: ¿crees que la AGI va a ocurrir? Dan: sí. ¿Es algo bueno? Dan: sí. Su definición de AGI — cualquier agente que tenga sentido económico mantener en marcha de forma continua, generando tokens activamente y completando tareas sin necesidad de nuevas instrucciones — es lo suficientemente precisa como para ser comprobable. Su razonamiento: incluso un sistema verdaderamente autónomo habrá sido construido para servir objetivos humanos; si no, nadie lo construiría. La preocupación de Brandon es que, una vez que los agentes continuos sean económicamente racionales, el argumento de los despidos masivos empiece a tener sentido. > *"Cualquier agente que nunca apagas — que tiene sentido económico mantener funcionando todo el tiempo, haciendo tareas de forma activa sin que tengas que volver a darle instrucciones."* ## [21:57] Los despidos por IA son una mentira Dan y Brandon analizan el caso ClickUp — un CEO que despidió públicamente a buena parte de su plantilla y lo atribuyó a la IA. La lectura de Dan: las empresas SaaS genéricas echan gente cuando tienen problemas o están sobredimensionadas, y luego culpan a la IA para cubrirse. Brandon añade que la réplica de Jensen Huang — "si tu respuesta al progreso es despedir gente, no eres un CEO muy creativo" — es interesada, pero probablemente cierta. El encuadre honesto: la IA transforma los flujos de trabajo en profundidad, lo que fuerza reorganizaciones a nivel de toda la empresa. Las compañías que se saltan ese trabajo y simplemente recortan plantilla están tomando el camino fácil. El keylogging de empleados por parte de Meta para recolectar datos de entrenamiento se menciona brevemente como una alternativa más creativa, aunque inquietante. > *"Sería muy escéptico de cualquiera que diga que va a eliminar todos los empleos o todo el trabajo del conocimiento."* ## [25:42] Adáptate a los modelos y todo irá bien Incluso en un escenario de AGI, la variable crítica es el juicio humano sobre *qué importa* — y lo que importa cambia constantemente, en parte porque la propia IA no deja de remodelar el mundo. Los trabajadores de atención al cliente en Omaha que desconfían de los chatbots, o las empresas que despiden al equipo de soporte y lo vuelven a contratar en silencio dos meses después, ilustran cuánto se rezaga la adopción real respecto al hype. La adopción tarda una generación en consolidarse; al final todos tendrán acceso a estas herramientas; quienes saldrán adelante son quienes sigan aprendiendo cada nuevo modelo a medida que salga. Dan cierra con su frase más clara: si te adaptas a los modelos, todo irá bien. > *"Si simplemente te adaptas a los modelos — cuando salgan nuevos modelos, aprende a usarlos para lo que tú haces, sea lo que sea — todo irá bien."* ## [35:30] Cómo usar la IA como editor de artículos de largo aliento Dan describe el proceso concreto con IA que hay detrás de "After Automation". Cada mañana monologaba el estado actual del argumento en Proof, luego le pasaba ese registro a Claude y le preguntaba: "¿Qué estoy intentando decir realmente?" Cuando los borradores superaron las 4.000 palabras, usó Codex para convertir la última versión en un podcast y lo escuchaba durante el trayecto al trabajo, detectando problemas de ritmo sin tener que mirar la pantalla. El texto pasó por cuatro o cinco reinicios completos antes de que el argumento encajara. Su conclusión: la IA no escribió el ensayo, pero hizo posible mantener toda la estructura de 8.000 palabras en la memoria de trabajo sin perder el hilo. > *"No podría haberlo escrito sin ella. Le pedía a Claude que tomara mi registro y me dijera: '¿Qué estoy intentando decir realmente?' Y me respondía cosas, y yo pensaba: 'Ah, eso es lo que intento decir.'"* ## Entidades - **Dan Shipper** (Persona): Cofundador y CEO de Every; presentador habitual de *AI & I*; aquí el entrevistado que habla sobre su ensayo "After Automation" - **Brandon Gell** (Persona): COO de Every; conduce este episodio entrevistando a Dan en un giro de formato - **Every** (Organización): Empresa de medios y software nativa en IA; ha crecido de 4 a 30 personas desde GPT-3 mientras automatiza intensamente; publica el podcast *AI & I* - **After Automation** (Concepto): Ensayo de 8.000 palabras de Dan Shipper que argumenta que la automatización con IA aumenta la demanda de trabajo humano experto al inundar los sectores con resultados casi correctos - **Brecha de competencia experta** (Concepto): La tesis de que la IA ofrece "la competencia experta de ayer" de forma barata pero siempre ligeramente desajustada, creando más necesidad de humanos que puedan cerrar esa brecha con la situación real - **AGI** (Concepto): Definida en este episodio como cualquier agente que tenga sentido económico mantener en marcha de forma continua sin necesidad de nuevas instrucciones; Dan cree que ocurrirá y que es positiva en términos netos - **Autonomía vs. agencia** (Concepto): La distinción de Brandon entre la IA ejecutando tareas abiertas sin supervisión (autonomía) y la IA con motivaciones propias (agencia); esto último no se está construyendo - **Proof** (Software): Herramienta de escritura que Dan usa para sus borradores diarios en voz; utilizada como bucle de retroalimentación con IA durante el desarrollo del ensayo - **Codex** (Software): Herramienta de OpenAI que Dan usó para convertir borradores del ensayo en formato podcast de audio para revisarlos durante el trayecto - **ClickUp** (Organización): Empresa SaaS cuyo CEO despidió públicamente a buena parte de la plantilla atribuyéndolo a la IA; usada como caso de estudio de despidos maquillados con IA

Noah Brier corre Claude Code en un mini PC de su sótano, sincronizado con su vault de Obsidian a través de una VPN de Tailscale, y hace pensamiento real, investigación y código para clientes desde su teléfono. La conversación aborda cómo construyó este stack, por qué impone restricciones estrictas de "modo de pensamiento" para evitar que el modelo redacte artefactos prematuramente, y su teoría más amplia de que la IA triunfa cuando se mete en los rincones y grietas de las organizaciones en lugar de exigirles que adopten nuevas estructuras. Dan Shipper y Noah también exploran qué significa realmente construir intuición sobre la IA y por qué preparar a los hijos para la IA tiene más que ver con enseñarles escepticismo epistémico que con vigilar el plagio. ## [00:00] La configuración de Claude Code de Noah Brier en un servidor casero Dan Shipper abre el episodio describiendo la configuración que hace que Noah valga la pena escuchar: un servidor doméstico en el sótano que corre Claude Code sobre un vault de Obsidian, accesible desde cualquier lugar a través del teléfono. Noah lo ha montado de modo que puede pensar, investigar, escribir y publicar código sin tener que sentarse ante un escritorio. > *"Armó un servidor en su sótano, le puso su vault de Obsidian y corre Claude Code encima para poder pensar, investigar, escribir e incluso publicar código directamente desde su teléfono."* ## [00:52] Introducción Dan y Noah se ponen al día, su primera conversación en unos cinco años. El recorrido de Noah abarca la estrategia de marca (cofundó Percolate), la consultoría de IA en Alephic y la conferencia BRXND.AI. Dan enfoca la entrevista en el stack práctico que Noah ha construido, no en la discusión abstracta sobre IA. > *"Me alegra tenerte aquí. Qué bueno poder charlar. Esta es nuestra primera entrevista en probablemente como 5 años."* ## [02:10] Cómo puedes hacer trabajo profundo desde tu teléfono Noah aclara desde el principio que su configuración es menos "vibe coding" y más trabajo del conocimiento estructurado. Abandonó Evernote por Obsidian porque los archivos markdown y las carpetas le dan algo sobre lo que Claude Code puede operar realmente. Su caso de uso principal de Claude Code es interactuar con sus notas, no generar código, y la extensión móvil de esa configuración ha cambiado fundamentalmente sus patrones de trabajo. > *"Mi uso número uno de Claude Code es usarlo como herramienta para interactuar con mis notas."* ## [05:30] Por qué Noah cree que Grok tiene el mejor modo de voz de IA Noah prefiere el modo de voz de Grok sobre los equivalentes de OpenAI y Gemini: Gemini no era suficientemente inteligente, y el antiguo modo de voz de GPT-4o le resultaba inutilizable. Lo usó en un viaje en solitario de cinco horas para trabajar en un artículo sobre los Transformers, conectándolo por Bluetooth y tratándolo como un podcast de investigación personal. La conversación saca a relucir una frustración compartida: los modelos de voz todavía no hacen bien el tool calling ni la investigación web, lo que limita su utilidad para trabajo intelectual serio. > *"Hice una sesión de una hora y realmente, fue de lejos la mejor explicación que he leído o escuchado sobre eso."* ## [11:11] Los detalles técnicos de la configuración Claude Code-Obsidian de Noah Noah recorre en pantalla su carpeta de Obsidian en vivo. Claude Code está en el directorio raíz de Obsidian, así puede acceder a todo el archivo de notas. Para una charla que está preparando para BRXND.AI, sobre el Simple Sabotage Field Manual de la Segunda Guerra Mundial y lo que dice sobre la burocracia en las grandes organizaciones, ha creado una carpeta de proyecto dentro de Obsidian con transcripciones de chats con ChatGPT, Claude y Grok, junto con artículos y PDFs. El trabajo de Claude en esta etapa no es escribir la charla sino ayudarle a pensar: extrae notas relevantes, sintetiza el progreso diario en un registro y hace preguntas clarificadoras. Noah establece restricciones de modo de pensamiento explícitamente en el frontmatter del CLAUDE.md del proyecto. > *"Estoy en modo de pensamiento, no en modo de escritura todavía. Hay cosas aquí donde específicamente le he dicho, creo que está en el frontmatter, donde le he dicho a Claude Code: no me ayudes a escribir nada ahora mismo."* ## [26:05] Usar un agente en Claude Code como 'compañero de pensamiento' Noah argumenta que la palabra "generativo" ha distorsionado la forma en que la gente usa la IA: todo el mundo se centra en su capacidad para producir artefactos, casi nadie habla de lo extraordinaria que es su capacidad de lectura. Mantiene un agente compañero de pensamiento con restricciones explícitas: "No crees esquemas, borradores ni ninguna versión de charlas o escritura." El agente anota preguntas, rastrea ideas emergentes y construye un registro continuo para que Noah pueda retomar exactamente donde lo dejó tras un descanso. Sigue un hilo desde la investigación profunda de ChatGPT sobre Wild Bill Donovan hasta una idea tentativa sobre cómo el paralelismo de la arquitectura transformer refleja la autonomía operacional de las fuerzas especiales. > *"Creo que en parte porque lo llamamos generativo, hay demasiado enfoque en su capacidad para escribir y no suficiente en su capacidad para leer."* ## [30:23] La teoría del Thomas' English Muffin de la IA según Noah El capítulo abre con la tesis de Noah sobre la burocracia: las grandes empresas no fracasan en adoptar software porque sean perezosas, sino porque históricamente el nuevo software exigía que las organizaciones se reestructuraran a su alrededor. La IA, argumenta, es diferente. Se mete en los rincones y grietas de cómo la gente ya trabaja, de ahí su metáfora del Thomas' English Muffin. Dan añade un ejemplo concreto de Every: dos productos construidos sobre stacks diferentes necesitaban compartir una solución de búsqueda de archivos, y Claude Code les permitió reutilizar la lógica sin imponer un framework común. La conversación se amplía a la idea de Noah sobre la "burocracia como positional encoding", una analogía a medio formar entre la arquitectura transformer y la jerarquía organizacional que todavía está elaborando antes de su charla. > *"Lo llamo mi Thomas' English Muffin theory de la IA, que es que se mete en los rincones y grietas."* ## [39:47] El espacio en blanco que aún queda por explorar en la IA Noah y Dan argumentan que la mayoría de los profesionales, incluidos los bien financiados, aún operan con intuiciones frágiles sobre lo que estos modelos realmente pueden hacer. El rompehielos de Noah en cada reunión con clientes es "¿cuál fue tu momento aha con la IA?" porque ese momento de no determinismo, hacer la misma pregunta dos veces y obtener respuestas diferentes, es genuinamente nuevo y requiere tiempo para internalizarse. Toma prestado el experimento de la bicicleta al revés de Destin Sandlin para ilustrar el punto: la intuición motora y la intuición conceptual son separadas, y no se puede atajar la construcción de intuición. Dan contraargumenta que los modelos de lenguaje podrían generar el vocabulario que nos falta para razonar sobre sistemas probabilísticos. > *"No estamos acostumbrados a usar cosas que, les haces la misma pregunta dos veces y dan respuestas diferentes."* ## [48:44] Cómo Noah prepara a sus hijos para la IA La hija de diez años de Noah construyó una aplicación de Papá Noel secreto con Claude que accidentalmente le enseñó modelado de datos: se dio cuenta de que necesitaba "grupos" en lugar de "adultos y niños" para generalizar la lógica. Esa historia ancla un argumento más amplio: el trabajo de los educadores no es prevenir el uso de la IA sino convencer a los estudiantes de que las habilidades subyacentes valen la pena aprenderlas. Noah está proponiendo un curso en NYU para el otoño de 2026 llamado "Code is Essay", y cree que la meta-habilidad relevante es el escepticismo epistémico: ser más desconfiado de la información que confirma tus supuestos previos, no menos. > *"En realidad no creo que tu trabajo sea enseñarles a escribir a estos niños porque eso es una búsqueda de toda la vida. Creo que tu trabajo es convencerlos de que vale la pena aprender a escribir."* ## [01:00:06] Cómo llevó su configuración de Claude Code al teléfono Noah hace una demostración en vivo del stack móvil completo: Termius (cliente SSH en iPhone), Tailscale VPN conectando al mini PC del sótano, Obsidian sincronizado vía GitHub privado, Claude Code corriendo en la terminal. Muestra cómo le pregunta a Claude "¿qué hay de nuevo en los últimos dos días?" y obtiene una síntesis de su actividad reciente en Obsidian. También arregló un enlace roto en su sitio de conferencias desde su teléfono: confirmó el bug, hizo que Claude enviara un PR, listo. Su tinkering actual se extiende a la herramienta CLI `llm` de Simon Willison y un script que renombra todos los archivos de attachments en su vault de Obsidian y reconstruye la tabla de enlaces. > *"Fui a sentarme afuera un rato y teníamos un proyecto que necesitaba entregarse a un cliente y había que hacer un pequeño cambio. Le dije a Claude Code exactamente dónde mirar, confirmé que el problema era lo que creía, y le pedí que enviara una solución, envió un PR y listo."* ## Personajes - **Dan Shipper** (Persona): CEO y cofundador de Every; conductor de la entrevista - **Noah Brier** (Persona): Cofundador de Percolate; fundador de la consultora de estrategia de IA Alephic; organizador de la conferencia BRXND.AI - **Every** (Organización): Empresa de medios y software que produce este podcast - **Alephic** (Organización): Consultora de estrategia de IA de Noah; trabaja con clientes Fortune 50 como Amazon, Meta y PayPal - **BRXND.AI** (Organización): Conferencia anual en la intersección de marketing e IA, organizada por Noah; edición 2025 en Nueva York el 18 de septiembre - **Claude Code** (Software): Herramienta de programación agéntica de Anthropic; central en el flujo de trabajo de segundo cerebro y móvil de Noah - **Obsidian** (Software): Aplicación de toma de notas basada en markdown; el almacén de conocimiento principal de Noah, organizado con el método PARA - **Tailscale** (Software): VPN de malla usada para conectar de forma segura el teléfono de Noah a su mini PC del sótano - **Termius** (Software): Cliente SSH para iOS que Noah usa para acceder a su servidor doméstico desde el teléfono - **Grok** (Software): Asistente de IA de xAI; Noah considera que su modo de voz es significativamente mejor que el de OpenAI y Gemini para investigación sustantiva - **Simple Sabotage Field Manual** (Concepto): Documento de la OSS de la Segunda Guerra Mundial que Noah republicó; usado como lente sobre la burocracia organizacional moderna en su charla de BRXND.AI - **Thomas' English Muffin theory** (Concepto): Metáfora de Noah sobre cómo la IA triunfa al integrarse en los flujos de trabajo organizacionales existentes en lugar de exigir reestructuración

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