When would you use graph RAG instead of standard vector RAG?
RAG · Pro 1
Short answer
Use it when questions depend on relationships across many documents or need a summary of a whole collection, such as 'what are the main themes in these reports'. Graph RAG extracts entities and relationships into a knowledge graph and retrieves along links. It costs far more to build, so try it only where vector RAG fails.
Deeper answer
Vector RAG finds chunks similar to the question, which works for most fact lookups. It struggles with global questions, because no single chunk holds the answer, and with multi-hop questions that chain facts.
Microsoft's GraphRAG popularised the approach. The extraction step uses many model calls over the whole corpus, so measure cost, check extraction quality, and plan how to update the graph when documents change.
What a strong answer covers
- Fits relationship or multi-hop questions
- Fits whole-collection summary questions
- Extracts entities and relationships into a graph
- Costly to build and update
- Use only where vector RAG fails
Answer it out loud in the mock interview about 1.5 min
Follow-ups to expect
- What is a multi-hop question?
- How would you keep the graph current?
Common mistakes
- Using graph RAG for simple lookups
- Ignoring indexing cost
Learn it
- How do you choose a chunk size for RAG?
- Your RAG app gives wrong answers. How do you find the cause?
- Design a RAG system that lets a law firm search and ask questions about 50,000 contracts.
- How would you detect hallucinations in a production RAG application?
- What are embeddings, and how are they used in search?
- How do you make a RAG system cite sources and stay grounded?
- What are hybrid search and reranking, and when do they help?
- How do you keep a RAG index up to date as documents change?
- How do metadata filters improve retrieval?
- How would you build RAG over documents full of charts and images?
- Your RAG system must handle scanned PDFs, tables and slides. How do you ingest them?
- How do you make sure users only get answers from documents they are allowed to see?
- What are query rewriting and HyDE, and when do they help?
- Your HR assistant keeps quoting last year's leave policy. How do you fix it?
- Tell me about a RAG system you built and how you measured whether it worked.
- Can you do retrieval for long documents without a vector database?
- What is RAG, and why would you use it?