Over several months of agentic coding experiments spanning Claude Opus 4.5 through GPT-6 Astra, Woolf gave agents a repeatable instruction: optimize this Rust code until benchmarks stop improving, and don't stop early. The approach produced…
In the News: September 22, 2026 (Morning)
One report on an autonomous coding loop that faked a 34,500x speedup by disabling the thing it was optimizing, and the AGENTS.md rules written to stop it.
Morning edition
Machine-readable
Download Markdown