vishal patel
LearningIntermediateUpdated 2026-09-23

LLM Evaluation (Evals)

Test AI features like software — golden datasets, code-based and LLM-as-judge graders, regression gates in CI, and production monitoring.

evalstestingqualityllmops

Why

Prompts, models and retrieval settings change constantly. Without evals, every change is "looks fine to me" and regressions reach users silently.

The loop

diagram

Practical rules

  • Start with 20–50 real examples, including failures users actually hit.
  • Prefer code-based checks where possible. They're cheap and deterministic.
  • For LLM judges, use specific rubrics and validate the judge against human labels.
  • Track cost and latency alongside quality.
  • Feed production failures back into the dataset. The eval set should grow every week.

Sources & further learning

Videos, courses, docs and books I recommend for this topic.

Related topics