We benchmarked a multilingual LLM judge on Spanish and Catalan text against human labels using Jev from TypeSafe. The result: keep rubric questions in English to maintain accuracy.
A practical guide to wiring LLM evaluation into CI/CD so quality regressions get caught before they ship, not after support tickets pile up.
How to measure whether an LLM application actually works, from traces and golden datasets through judge calibration, CI regression gating, and production monitoring.
