HN Explorer

kolesnikov-arch

↗ HN profile [ 33.1 ms ] full profile
Karma
0
Created
()
Submissions
0
  1. Show HN: I pre-registered the rules,then measured AI agents shipping wrong fixes (github.com)

    A controlled comparison of ungated vs gated AI patching on SWE-bench Lite. Full methodology, results, and reproducible scripts in the repo. Breakthrough for flash models.

    1 points 4