Eval maturity test
Where are you on the ladder?
Answer up to 6 questions to figure out where you're up to, and see what the next level takes.
The five levels
Each level builds on the one before it. The test finds your rung.
L1Eyeballing. You read traces by hand.
L2Labelled. You have rated a sample against a scoring rubric.
L3Calibrated. Judges continuously updated to agree with you, and you know the rate.
L4Validated. Evals are proven to predict a business metric.
L5Diagnosed. Failures traced to cause, changes measured.