techniques

Eval (evaluation)

A test for an AI system: you give it an input, then apply grading logic to its output to measure success. Because model outputs vary between runs, multiple trials are often run to get more consistent results. Think of a driving test, where the examiner sets a task and scores the result.

See it in practice: Evals: How Do You Know It's Working?

Want to learn more about AI?

Peter Saddington has trained 17,000+ people on agile and AI. Let’s talk.

Work with Peter