AI evals for product teams

An eval tells you whether your AI product is good. These articles cover how to build one as a product team: which judges to use, where the rubric and test cases come from before you have users, and how to check that an automated judge agrees with the people you're building for.

What a thumbs up can't tell you about your AI product

What a thumbs up can't tell you about your AI product

September 29, 2026

9 mins

LLM as a judge, and whether it scores what your users score

LLM as a judge, and whether it scores what your users score

September 29, 2026

9 mins

Users as your golden dataset: how to build evals before launch

Users as your golden dataset: how to build evals before launch

September 29, 2026

5 mins

Evals for product managers, before and after launch

Evals for product managers, before and after launch

September 29, 2026

11 mins

LLM as a judge vs human evaluation

LLM as a judge vs human evaluation

September 29, 2026

8 mins

Jev-as-a-judge vs LLM-as-a-judge for product managers

Jev-as-a-judge vs LLM-as-a-judge for product managers

September 29, 2026

8 mins

Types of evals for AI products, sorted by judge

Types of evals for AI products, sorted by judge

September 29, 2026

7 mins

What are AI evals? Five types, and the one most teams are missing

What are AI evals? Five types, and the one most teams are missing

September 21, 2026

10 mins

Evaluation rubrics: defining 'good' for your AI product

Evaluation rubrics: defining 'good' for your AI product

September 1, 2026

5 mins