Evaluation

Without a frozen eval set, every prompt, RAG, or model change is a guess — evaluation is how you know the system actually got better.

Offline eval is a frozen test set in CI; online eval is live users. Ship with offline gates, confirm in production — real queries always drift.

Human evaluation with a written rubric is the gold standard for open-ended quality — use it to build gold sets and calibrate judges, not to grade every reply.

LLM-as-a-judge grades outputs against a rubric at scale. Calibrate it on humans, and watch verbosity, position, and self-preference bias.

Public leaderboards pick models; your private gold set and CI tests pick whether your product got worse. Never train on the eval set.

Hallucination eval checks groundedness vs context, not "sounds true"; safety eval checks refusals, leaks, and overconfidence. Score them separately.

Production AI is a tradeoff of quality, latency, cost, and reliability. Optimize dollars per successful task under an SLO — not always the smartest model.