Add one risk-based AI eval slice to a current test suite. Use this when an AI workflow has a known gap such as stale retrieval, policy-sensitive answers, prompt injection, low-confidence classification, or missing source grounding. I will add [number] eval cases for [AI workflow risk] by [date]. Source of truth: [policy, fixture, document, dataset]. Oracle: [required facts, forbidden claims, relation, rubric, or human review lane]. Owner: [name]. I will check the next run for pass/fail and update the release note. In 2 days, check whether the eval slice exists, has an oracle, and appeared in at least one run report.…
Read the full lesson
Sign up free: one personalized lesson every day, matched to your role and goals.
Already have an account? Sign in