These are GenAI questions of the kind Amazon actually asks — the patterns reported from Amazon's GenAI and AI-engineering rounds, where design trade-offs, scale and safety matter as much as definitions. My advice: treat this page as a mock interview. Say every answer out loud before revealing it — GenAI rounds are conversations, not written tests. If an answer surprises you, the lesson behind it is linked at the bottom.
Amazon GenAI concept questions
Design a product-review summarizer that shows 'what customers say' on every product page, for millions of products. What are the key design decisions?
Asked in
Mechanistically, WHY does RAG reduce hallucinations — and name two ways a RAG system can still hallucinate.
Asked in
In tool/function calling, what security principles do you apply before letting an LLM's tool requests touch production systems?
Asked in
Amazon GenAI applied & hands-on questions
Your service makes LLM API calls that occasionally fail with rate-limit (429) errors. Write a Python wrapper that retries with exponential backoff and jitter, giving up after 4 attempts. Why is jitter included?
Asked in
How to use this page: Amazon rarely asks a question you've never seen — they ask standard GenAI concepts and then push one level deeper ("why?", "what would you do if..."). Master the concept in the GenAI course lessons, and the follow-up stops being scary.
Keep practising: How LLMs Work, Prompting Basics, RAG and Fine-tuning vs RAG cover what most Amazon GenAI rounds test.