All Lessons

What is an eval (evaluation) in the context of AI applications?

1 / 3
advanced 30 min

Testing AI Systems

Loading lesson content...

A team deploys a prompt change that passes all their evals but causes a 30% increase in bad outputs in production. What's the issue?

1 / 2

A team has 5000 test cases for their evals. Running all of them takes 2 hours, blocking rapid iteration. How to maintain quality while speeding up iteration?

1 / 1