swyx 🐣

They say you can’t go home again. Pretty true, but not entirely. It was fascinating to return to @NotionHQ to find many of the same hearts & minds at work (@ivanhzhao @akothari @simonlast & @davidtibbies among them) - but in entirely new ways for the AI era. Thrilled to document this shift alongside the magical @zebriez for @colossusmag.

🆕 The Full Story of Notion AI latent.space/p/notion We're so excited to chat with @simonlast and @sarahmsachs about Notion's "Token Town" - the crack team of AI Engineers and Model Behavior Engineers entrusted with building AI for Silicon Valley's most beloved knowledge work collaboration platform - and their latest launch of Custom Agents! We talked: • The full history of the 5 major rebuilds of Notion AI — and the key lessons from each • How to eval agent *usefulness* not just correctness • MCP vs CLI pros and cons • What "work" looks like when agents are coworkers — why they build for the "top of the class" rather than dumb down AI for everyone • Simon's take on the ideal "software factory" of the future and so much more! Timestamps: 00:00:00 Introduction and launching Notion Custom Agents 00:01:17 Why Notion rebuilt agents four or five times 00:03:35 Building for where models are going, not just where they are 00:05:32 The Agent Lab thesis, wrappers, and product intuition 00:08:07 User journeys, leadership, and low-ego AI teams 00:13:16 The Simon Vortex, hackathons, and bringing security in early 00:16:39 Team structure, demos over memos, and building for agents 00:20:25 Evals, Notion’s Last Exam, and the Model Behavior Engineer role 00:27:37 Evals as an agent harness and the changing role of software engineers 00:30:42 The software factory: specs, verification, and agent workflows 00:32:18 Live demo: a custom agent for coworking space applications 00:35:08 Composing agents, manager agents, and memory as pages 00:38:15 Notion Mail, Gmail, native integrations, and tools 00:39:43 MCP vs CLI and the cost of capability 00:44:13 When Notion uses MCP vs building its own integrations 00:47:43 The history of Notion’s agent harness rebuilds 00:55:35 Power users, public tools, and the setup agent 00:58:01 Self-fixing agents, permissions, and “flippy” 01:01:13 Pricing, credits, and choosing the right model automatically 01:09:01 Why Notion isn’t training its own frontier model 01:14:07 Retrieval, ranking, and search built for agents 01:17:27 Meeting Notes as data capture and workflow automation 01:21:18 Wearables, hardware, and Notion as the system of record 01:23:45 Outro

The only 4 jobs that will remain at tech companies. Credits: @yrechtman

In @steipete's latest State of the Claw, he gives an update on 5 months of @OpenClaw and some behind the scenes on what it's like maintaining the fastest growing open source of all time: youtube.com/watch?v=zgNvts… eg: - 60x more security reports than curl - a "Bullshit Taxonomy" of illegitimate reports - Nation State attacks - 12%-20% of skills contributions malicious - contributors burning multiple Codex Pro per day - academic FUD Agents are both the product AND the attack vector. @simonw's Lethal Trifecta is not solved. Come for Pete's recommendations, what OpenClaw is doing on security, OpenClaw Foundation roadmap, and then subsequence audience Q&A with @swyx on taste, dreaming, and OpenAI.

🆕 Harness Engineering: How to Build Software When Humans Steer, Agents Execute youtube.com/watch?v=am_oeA… @_lopopolo is one of the emerging class of token billionaires at @OpenAI, and is unapologetically exploring how you can get to 5 -> 50 -> 5000 agents working for you 24/7. Special Q&A moderated by @vibhuuuus.