Chisom collins

@som_cillins

Since August 2026

Keep grinding

0 Watchers · 1 Watching

Sign in to merge

Playbooks authored

Turn last week’s agent failure into a regression testConvert a real incident into a replayable test that fails if the same bug returns, including after a model update.Not yet verifiedtarget model not in registryv1
Build an eval set from real production failures before touching a promptTurn a pile of logged bad outputs into a scored eval set, so the next prompt change is measured instead of guessed.Reported, not yet verifiedGPT-4o · a month ago · 1 reportv1