These are Prompt Engineering questions of the kind Amazon actually asks — the patterns reported from Amazon's AI-engineering rounds, where design trade-offs, scale and failure modes matter as much as definitions. Treat this page as a mock interview: say every answer out loud before revealing it. If one surprises you, the lesson behind it is linked at the bottom.
Amazon Prompt Engineering concept questions
You have a 200k-token context window. Why not just put all retrieved documents in the prompt?
Asked in
Amazon Prompt Engineering applied & hands-on questions
Design the prompting for a feature that answers customer questions about a product using its description, specifications and existing customer reviews. Cover context assembly, the prompt itself, and what you would evaluate.
Asked in
A working feature costs too much. Its prompt: 900-line system prompt, 12 few-shot examples, all 20 retrieved passages, full chat history, and a request timestamp at the top. Quality must not drop. What do you change, and in what order?
Asked in
How to use this page: Amazon rarely asks something you've never seen — they ask a standard Prompt Engineering concept and then push one level deeper ("why?", "what would you do if..."). Master the concept in the Prompt Engineering course lessons, and the follow-up stops being scary.
Keep practising: Anatomy of a Prompt, Zero-shot & Few-shot, Chain-of-Thought and Prompt Injection & Safety cover what most Amazon Prompt Engineering rounds test.