Skip to content
Observability

Evals

How to know a change made your AI system better, before your users find out it did not.

Intermediate · 6 steps

Step 1 of 6· Illustrative golden set, replies and scores; the gate logic and the swap test follow the cited practice

in one minute

An eval is a test suite for an AI system. You collect real cases, each with an input and a way to check the answer, and run every version of the system against them. Some checks are plain code: did it quote the right price, call the right tool, keep card numbers out of the reply? Open-ended answers, such as tone or policy explanations, are graded by a second model against a written rubric. Comparing versions case by case shows what a change fixed and what it broke. A release gate then turns the results into a decision, so a better average cannot hide a broken safety rule.

Why it matters for your product

Without evals, every prompt tweak or model upgrade is a guess, and regressions reach customers first. A set of real cases with automated checks runs on every change, the way unit tests do for ordinary code. Keep must-pass rules separate from the average, check your judge against human labels, and add every real failure you see as a new case. The discipline matters more than any particular tool.

For engineersShow the maths

pass rate = cases passed ÷ cases run

The headline number: the share of golden-set cases whose check says pass.

Worked example: v1 passes 19 of 24 = 79.2%; v2 passes 21 of 24 = 87.5%, a rise of 8.3 points.

net change = cases fixed − cases broken

The pass rate only shows the net. The case-by-case diff shows both sides, and a single broken must-pass case outweighs any net gain.

Worked example: v2 fixed 4 cases and broke 2: net +2, with the safety case X3 among the broken.

judge agreement = cases where judge and humans agree ÷ cases both graded

How often the judge reaches the same verdict as people on a sample you have labelled by hand.

Worked example: 42 of 50 cases agree, so agreement is 84%.

where it stops working

A golden set measures only what it contains: a single held-out score tends to overestimate how well a model works in the wild, and if test cases leak into a model's training data, scores rise without any real improvement. With a few dozen cases, small differences between versions are mostly noise, which is why eval results need error bars.

Key terms

The words you will hear

Golden set

A fixed collection of representative cases, each with an input and a way to check the output.
Where it came from

Primary sources

  1. 01Judging LLM-as-a-Judge with MT-Bench and Chatbot ArenaZheng et al. · 2023
  2. 02Define success criteria and build evaluationsAnthropic · 2026
  3. 03Evaluating Large Language Models Trained on CodeChen et al. · 2021
  4. 04Beyond Accuracy: Behavioral Testing of NLP Models with CheckListRibeiro et al. · 2020
  5. 05NLP Evaluation in trouble: On the Need to Measure LLM Data Contamination for each BenchmarkSainz et al. · 2023
  6. 06Adding Error Bars to Evals: A Statistical Approach to Language Model EvaluationsMiller · 2024

Want this working on your data?

We design and build the systems these ideas power: retrieval, agents, voice and the models behind them. Start with a free discovery call.