10 questions on resolving instruction conflicts, designing a chain, validating structured output, building templates, planning long-context tests, comparing prompt variants, computing task metrics, estimating usage cost, mocking provider responses and designing regression cases.
One question at a time
Move with Previous and Next, and jump around with the question palette.
Pass mark
You need 6 of 10 marks to pass.
Timed
You get 15 minutes. The exam submits automatically when time runs out.
10 questions, 1 mark each. Pass mark: 6. Some questions have more than one correct answer, and some ask you to type a short answer. Prompts and model replies shown in questions are invented examples for practice: judge them with the checks the tutorial teaches, not by how fluent they sound. Code in questions is Python 3.12 and needs no API key.
Sign in to join the discussion.