Every resource founders actually use, in one library.

What founders use most, ranked by how many answers lean on it.

1. Video · How Founders Use AI Simon Willison on Industry's Tardy Response to Prompt Injection Simon Willison
2. A current-year synthesis tying the main threats to a defense stack. airia.com Article · How Founders Use AI AI Security in 2026: Prompt Injection, the Lethal Trifecta, and How to Defend Airia
3. Anthropic's launch post explaining the problem MCP was made to solve. anthropic.com Article · How Founders Use AI Introducing the Model Context Protocol Anthropic
4. The official introduction: what MCP connects, the USB-C comparison, and how clients and servers fit together. modelcontextprotocol.io Guide · How Founders Use AI What is the Model Context Protocol (MCP)? Model Context Protocol
5. Video · How Founders Use AI Model Context Protocol (MCP), Clearly Explained (Why it Matters) Builders Central
6. The contrarian essay from the Devin team that reset the industry's defaults. cognition.com Article · How Founders Use AI Don't Build Multi-Agents Walden Yan (Cognition)
7. The thread version of the argument, with the mistakes he keeps seeing. x.com Thread · How Founders Use AI Walden Yan: principles for building agents Walden Yan
8. The honest update: some multi-agent setups now work, most sexy ideas still don't. x.com Thread · How Founders Use AI Walden Yan: a year later, what changed Walden Yan
9. The follow-up naming the specific multi-agent patterns that survived production. cognition.com Article · How Founders Use AI Multi-Agents: What's Actually Working Cognition
10. The other side of the debate: a 90% performance gain, at 15x the tokens. anthropic.com Article · How Founders Use AI How we built our multi-agent research system Anthropic
11. Anthropic's own decision rules for graduating beyond a single agent. claude.com Article · How Founders Use AI When to use multi-agent systems (and when not to) Anthropic
12. Shows why Cognition and Anthropic are both right about different problems. philschmid.de Article · How Founders Use AI Single vs Multi-Agent System? Philipp Schmid (Google DeepMind)
13. The architecture broken down with diagrams a non-researcher can follow. blog.bytebytego.com Article · How Founders Use AI How Anthropic Built a Multi-Agent Research System ByteByteGo
14. 200+ analyzed tasks and 14 failure modes, the evidence behind the caution. arxiv.org Article · How Founders Use AI Why Do Multi-Agent LLM Systems Fail? UC Berkeley (MAST)
15. The open dataset and annotator to diagnose your own multi-agent failures. github.com Tool · How Founders Use AI MAST: Multi-Agent System Failure Taxonomy UC Berkeley Sky Computing Lab
16. The digestible summary of where multi-agent systems break. sky.cs.berkeley.edu Article · How Founders Use AI MAST project page UC Berkeley Sky Computing Lab
17. A neutral recap of the whole debate if you want it in ten minutes. ctol.digital Article · How Founders Use AI AI Leaders Clash Over Agent Architecture CTOL Digital
18. Frames the choice as a coordination-cost calculation, which is the right lens. cognilium.ai Article · How Founders Use AI Multi-Agent vs Single Agent: When It's Worth the Cost Cognilium
19. Extracts the engineering lessons hiding inside Anthropic's post. llmmultiagents.com Article · How Founders Use AI The Art of Multi-Agent Collaboration: Reflections on Anthropic's Research System LLM Multi Agents
20. Maps each MAST failure mode to a practical fix. futureagi.substack.com Article · How Founders Use AI Why do multi-agent LLM systems fail (and how to fix) Future AGI
21. The paper's findings retold for people who won't read a paper. thegrigorian.medium.com Article · How Founders Use AI Why Do Multi-Agent LLM Systems Fail? (explainer) Anna Alexandra Grigoryan
22. The case study written for someone deciding whether to copy the pattern. zenml.io Article · How Founders Use AI Anthropic's multi-agent research system (LLMOps case study) ZenML
23. Where coding agents go next, from the person who built the category leader. lennysnewsletter.com Podcast · How Founders Use AI Head of Claude Code: What happens after coding is solved | Boris Cherny Lenny's Podcast (Boris Cherny)
24. The engineering story, including the 80-90% self-written stat. open.spotify.com Podcast · How Founders Use AI Building Claude Code with Boris Cherny The Pragmatic Engineer (Gergely Orosz)

Show the next 24