Anthropic's Fermat Run Failed on Lost Agent State, Solved With a Proof DAG
inductionheads · x · 2026-09-06
Chi Wang cites Anthropic's retrospective: its first Fermat run failed not on math, but because dozens of Claude agents lost track of shared project state.
The team only succeeded after adopting Prove2Me's DAG of statements and proofs. The takeaway: at multi-agent scale, "memory" must be external, structured infrastructure rather than prompts. The retweeter muses what if every scientific problem had a DAG.
More from coding & agent
- Dev demos AI models handling 3D objects and animations surprisingly well — chongdashu · 2026-10-23
- Jason Kneen burns 104.56B tokens and $24.7K in 304 days of AI coding on Codex and Claude Code — jasonkneen · 2026-10-23
- Typosquatted Blender MCP Repo Hit 958 Stars in Days and Silently Ran Malware — Promptmethus · 2026-10-23
- Benchmarking 56 fresh dependency bugs: MCP routing lifts Claude Code accuracy 68% to 91% — Available_Side_7134 · 2026-10-23
- Swapping only the eval harness moves results 30%: designing agent evals like a scientist — AI Engineer · 2026-10-23
- Med student builds a whole-library RAG with Cognee; single-book ingestion still freezes his machine — shrike293103 · 2026-10-23