These are RAG questions of the kind Infosys actually asks — the patterns that keep coming back in Infosys's AI project interviews, where what matters is applying RAG sensibly for clients rather than research depth. 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.
Infosys RAG concept questions
What is an embedding, and why must the same embedding model be used for both documents and queries?
Asked in
A client insists on fine-tuning a model on their 800-page product catalogue so the bot 'knows the products'. How do you respond?
Asked in
Your RAG bot has been quoting a policy that was withdrawn two months ago. What went wrong and how do you prevent it?
Asked in
Infosys RAG applied & hands-on questions
Users ask "how many privilege leaves can I carry forward?" and get "I don't have that information", although the handbook covers it under 'Earned Leave — Accumulation'. Diagnose and fix.
Asked in
How to use this page: Infosys rarely asks something you've never seen — they ask a standard RAG concept and then push one level deeper ("why?", "what would you do if..."). Master the concept in the RAG course lessons, and the follow-up stops being scary.
Keep practising: What Is RAG?, Chunking, Retrieval Techniques and Evaluating RAG cover what most Infosys RAG rounds test — and the Infosys exam guide covers the rest of their selection process.