Interview Prep
RAG Interview Questions
Retrieval-Augmented Generation — chunking, embeddings, vector search and grounded answers.
10
Intermediate
What is RAG and when would you use it instead of fine-tuning?
RAG retrieves relevant documents at query time and gives them to the model as context. Use it when knowledge changes often or must be cited; fine-tune to change style or behaviour.
Key Points:
- RAG is cheaper to update than retraining.
11
Intermediate
How do you choose chunk size?
Balance context and precision: small chunks retrieve precisely but lose context; large chunks keep context but dilute relevance. Start around a few hundred tokens with some overlap and test retrieval quality.
Key Points:
- Chunk on document structure (headings) when you can.
12
Expert
What is re-ranking?
A second pass that scores the top retrieved chunks with a more accurate (slower) model so the best context goes into the prompt.
Key Points:
- Retrieve wide, re-rank narrow.