Technical interview guide

Retrieval-Augmented Generation (RAG)

Grounding an LLM's output in retrieved external documents instead of relying purely on its trained-in knowledge.

Read: 45 min
Practice MCQs: 25
Interview QA: 25
Edition: v4
Editorial status: Reviewed
Relevant for: AI EngineerAgentic AI EngineerForward Deployed Engineer (FDE)

Scope: RAG, DPR, Sentence-BERT, ColBERT, FAISS, HNSW, BEIR, MTEB, long-context, HyDE, RAGAS, NIST AI RMF, and OWASP prompt-injection references reviewed 2026-09-04.

Interview QA

Treat each question like a live interview question: answer out loud first (structure, assumptions, tradeoffs), then open the model answer to spot gaps and rehearse a tighter follow-up.

Curated: 2026-10-05 · Written: 2026-09-10 · Reviewed: 2026-10-03

QA-1

Decide whether a product needs RAG, fine-tuning, tools, or a base model.

QA-2

Design a governed ingestion pipeline.

QA-3

Choose and evaluate chunking for mixed documents.

QA-4

Migrate an embedding model without corrupting retrieval.

QA-5

Tune and validate an ANN index.

QA-6

Design hybrid retrieval and score fusion.

QA-7

Implement tenant and document ACL enforcement in RAG.

QA-8

Design safe query expansion and decomposition.

QA-9

Add a cross-encoder or late-interaction reranker.

QA-10

Choose dynamic context selection.

QA-11

Defend a RAG system against indirect prompt injection.

QA-12

Implement and evaluate trustworthy citations.

QA-13

Design RAG abstention and fallback behavior.

QA-14

Design source precedence and conflict handling.

QA-15

Build an offline retrieval evaluation set.

QA-16

Choose retrieval metrics for different RAG tasks.

QA-17

Design answer and citation evaluation.

QA-18

Design deletion and permission-revocation propagation.

QA-19

Define RAG freshness and reconciliation SLOs.

QA-20

Design safe retrieval and answer caching.

QA-21

Design RAG degradation and disaster recovery.

QA-22

Set RAG latency and cost budgets.

QA-23

Design online monitoring for a RAG product.

QA-24

Plan a safe RAG pipeline release.

QA-25