Field notes

Production AI agent reliability, written from the engagements.

Eval setups, failure modes, observability patterns, and the things we learn auditing production AI agents. Written for technical founders and AI engineering teams.

Why 74% of AI customer agents get rolled back — and what the 26% do differently
May 19, 2026
Cursor's agent deleted a 250GB production server from the prompt 'sync' — the audit checklist that would have caught it
May 19, 2026
Anthropic's Dreaming is brilliant. Putting your evals inside your model provider is not.
May 17, 2026