Re:Show HN: Beval – Simple evaluations for your AI product
by warwickmcintosh on Sunday March 29, 2026 @05:40PM
(#47567636)
LLM as judge drifts in weird ways if you don't have ground truth to calibrate against. Good that you've got that built in. Would love to see eval stability tracking over time though, same prompt different day sometimes gives different scores.
[ Parent
| on HN ]
There are no comments on this comment yet.