All 44 questions
How does semantic search differ from keyword search, and when would you use it?
Asked in 8 of 191 interviews · last asked Nov 6 · Concept · Junior
Show answer
Semantic search finds results by meaning, often using embeddings, while keyword search mainly matches words. Use semantic search when users express the same idea differently from the documents. Use keyword search for exact names, codes, or phrases, and combine both when you need both kinds of matching.
- Semantic search helps with natural-language questions, synonyms, and paraphrases.
- Keyword search is useful when the exact wording matters.
- Similarity search finds nearby items under a chosen metric; semantic search aims to match meaning.
- Hybrid search combines keyword and semantic results.
What the interviewer checks: The interviewer checks whether you can choose a search method based on user questions and document content.
Other ways it was asked
- “What do you mean by semantic search?”
- “What do you mean by semantic search, and how is it different from normal search?”
- “What is semantic search, and when would you use it?”
How would you implement hybrid search combining vector similarity and keyword search?
Asked in 2 of 191 interviews · last asked Jul 16 · Design · Senior
Show answer
I would run vector search and keyword search, such as BM25, on the same query, usually in parallel. Then I would merge results by document or chunk ID and combine their ranks using reciprocal rank fusion. Finally, I would optionally rerank the merged candidates and return the top results.
- Vector search finds related meaning; keyword search catches exact terms, names, and codes.
- Rank fusion avoids directly adding scores from different scales.
- Apply the same access filters to both searches and remove duplicate results.
- Tune candidate counts and fusion weights using representative queries and relevance labels.
What the interviewer checks: The interviewer checks whether you can combine complementary retrieval methods while handling incompatible scores, duplicates, and access filters.
Other ways it was asked
- “How would you implement hybrid search for a user query, combining semantic similarity search with keyword search?”
- “How would you build a hybrid search system that combines vector retrieval with traditional keyword search?”
What is query rewriting in retrieval, how does it work, and when would you use it?
Asked in 2 of 191 interviews · last asked Sep 9 · Concept · Mid-level
Show answer
Query rewriting turns a user's question into a clearer search query before retrieval. Rules or an LLM can add context, expand terms, or split a complex question into smaller queries. In vector search, the rewritten text is embedded and used to find documents. Use it for vague questions, follow-up questions, or wording that differs from your documents.
- Keep the user's meaning; do not add unsupported assumptions.
- Rewriting changes the search input, not the stored documents.
- Multiple queries can improve coverage, but add cost and may retrieve irrelevant results.
- Check whether rewriting improves retrieval quality; clear queries may not need it.
CodePrevious question: What is our refund policy? Follow-up: Does it cover sale items? Rewritten query: What is our refund policy for sale items?What the interviewer checks: The interviewer checks whether you understand how rewriting improves retrieval and when it may hurt.
Other ways it was asked
- “What is query rewriting in retrieval with a vector database, and when would you use it?”
- “How does query rewriting work?”
What retrieval methods would you use to fetch relevant chunks from a vector database?
Asked in 2 of 191 interviews · last asked Nov 17 · Concept · Mid-level
Other ways it was asked
- “What retrieval methods have you used to fetch the right chunks from a vector database during development?”
- “How do you retrieve the most relevant chunks from a vector database?”
What is BM25?
Asked in 2 of 191 interviews · last asked Oct 28 · Concept · Junior
Other ways it was asked
- “How does BM25 work?”
How do you choose between hybrid search and embedding-based search in a RAG system?
Asked once · last asked Mar 16 · Concept · Senior
Other ways it was asked
- “How would you combine BM25 lexical search with embedding-based search in a RAG system?”
What is BM25, and when would you use it for lexical search?
Asked once · last asked Mar 16 · Concept · Mid-level
What is the practical difference between BM25 and embeddings?
Asked once · last asked Mar 16 · Concept · Mid-level
How do you choose a RAG retrieval strategy, and what trade-offs guide your choice?
Asked once · last asked Apr 20 · Concept · Senior
How would you ensure an AI assistant retrieves the correct procedures when different sites have different operating processes?
Asked once · last asked May 13 · Scenario · Senior
If job descriptions change by position, how would you train the agent to match each resume to the correct job description at runtime?
Asked once · last asked Jun 9 · Design · Senior
How would you leverage vector embeddings to address cross-lingual semantic search in a multilingual search platform, and what challenges might you encounter during implementation?
Asked once · last asked Jul 16 · Design · Senior
How would you retrieve past tickets relevant to a new ticket's description?
Asked once · last asked Aug 7 · Design · Senior
Other ways it was asked
- “How would you retrieve relevant information from an AI agent's memory of past tickets?”
- “How would you find past tickets that are relevant to a new ticket's description?”
For a chatbot that answers questions using a PDF knowledge base, how would you find the most relevant PDF for a user's question?
Asked once · last asked Aug 7 · Scenario · Junior
How would you provide an LLM with information about relationships within a large dataset without overwhelming its context window?
Asked once · last asked Aug 7 · Design · Senior
Other ways it was asked
- “How would you give an LLM the interconnectedness of a large dataset without overwhelming it?”
How would you design retrieval for unstructured data with multi-hop relationships (A → B → C → D → Z) to target about 90% accuracy?
Asked once · last asked Aug 20 · Scenario · Senior
What is reranking in a RAG system, and what is vector ranking?
Asked once · last asked Aug 22 · Concept · Mid-level
How do you optimize your chunking strategy, embedding generation, or vector retrieval?
Asked once · last asked Aug 25 · Scenario · Senior
How does a RAG pipeline use indexed metadata to retrieve relevant content and generate a response?
Asked once · last asked Sep 5 · Concept · Senior
How would you combine BM25 with k-nearest-neighbor vector search for semantic video search?
Asked once · last asked Sep 7 · Concept · Senior
How would you decide whether each user query should trigger retrieval in a RAG chatbot?
Asked once · last asked Sep 10 · Design · Senior
How would you retrieve related information from multiple documents to answer a single query in a RAG system?
Asked once · last asked Sep 10 · Design · Senior
How would you improve retrieval when a system returns text that is lexically similar but semantically irrelevant?
Asked once · last asked Sep 18 · Scenario · Senior
How can retrieval filtering or confidence thresholding improve retrieval from external data sources?
Asked once · last asked Sep 22 · Concept · Senior
How would you rerank or refine the top results returned by Pinecone before passing them to an LLM, considering scoring adjustments, filtering, hybrid search, or LLM-based reranking?
Asked once · last asked Sep 23 · Scenario · Senior
Other ways it was asked
- “How did you handle re-ranking or refining the top results returned by Pinecone before passing them to the LLM? Did you use scoring adjustments, filtering, hybrid search, or LLM-based reranking?”
How does Azure implement hybrid search that combines vector search with BM25?
Asked once · last asked Oct 2 · Concept · Senior
How are relevance scores calculated in a RAG search system, and what is a vector database?
Asked once · last asked Oct 22 · Concept · Senior
What is an effective way to search hundreds of documents for a specific policy?
Asked once · last asked Nov 18 · Design · Mid-level
Other ways it was asked
- “How can you search hundreds of documents for a policy cost-effectively?”
- “How would you build a prompt or retrieval approach to find a specific policy across hundreds of documents?”
- “How would you optimize document search for a specific policy to reduce costs?”
How does RAG retrieve information when a query contains an exact term?
Asked once · last asked Nov 20 · Concept · Mid-level
Other ways it was asked
- “How does RAG retrieve information when a query uses an imprecise term?”
- “Can RAG retrieve relevant information when a query uses an imprecise term?”
Why might you use GraphRAG in a RAG system?
Asked once · last asked Nov 13 · Concept · Mid-level
Other ways it was asked
- “What is the architecture of a RAG system, and why might you use GraphRAG?”
Why do we need a reranker if we already have retrieval?
Asked once · last asked Oct 26 · Concept · Senior
What reranker models can you use in a RAG system?
Asked once · last asked Oct 26 · Concept · Mid-level
How does reciprocal rank fusion (RRF) work?
Asked once · last asked Oct 28 · Concept · Mid-level
What role can ranking methods play in retrieving relevant data in a RAG workflow?
Asked once · last asked Nov 4 · Concept · Mid-level
How would you handle a user asking about a disease using a different term than the one in the vector database, so the system returns the correct answer instead of a wrong one?
Asked once · last asked Nov 5 · Scenario · Mid-level
How would you choose a similarity metric and top-k value for a FAISS retriever?
Asked once · last asked Nov 13 · Scenario · Mid-level
How would you refine the reranking strategy if the top retrieved results still include irrelevant or low-quality passages?
Asked once · last asked Nov 15 · Scenario · Senior
How would you balance precision and recall when retrieval needs to support both comprehensive coverage and avoiding hallucinations in a live production environment?
Asked once · last asked Nov 15 · Scenario · Senior
What criteria would you use to set a retrieval relevance threshold, and how would you adjust it if the model began missing key details or hallucinating irrelevant content?
Asked once · last asked Nov 15 · Scenario · Senior
What is hybrid retrieval, and when would you use it?
Asked once · last asked Nov 18 · Concept · Mid-level
What similarity measure can be used between a query and documents in the probabilistic model?
Asked once · last asked Jul 31 · Concept · Junior
What is metasearch?
Asked once · last asked Apr 27 · Concept · Junior
What is re-ranking, and how is it used in information retrieval?
Asked once · last asked Sep 28 · Concept · Junior
What strategy would you choose for implementing semantic search, and why?
Asked once · last asked Nov 7 · Concept · Mid-level
Other ways it was asked
- “Which strategy did you choose for implementing semantic search?”
More RAG interview questions
- All RAG interview questions
- Embeddings and Semantic Similarity Interview Questions
- RAG Evaluation and Debugging Interview Questions
- RAG System Design and Agent Interview Questions
How this list is made
Every question here was asked by a real interviewer in a real job interview that a candidate ran MeetAssist in. After the call, the interviewer's questions are pulled from the transcript. Practice sessions, tests, exams and anything that was not a job interview are dropped.
Questions are rewritten so they make sense on their own: names of people, employers and clients are removed, and speech-to-text mistakes are fixed. Small talk, logistics and questions about one candidate's own résumé are left out. Rephrasings of the same question are merged, and the count is the number of different interviews it came up in.
The interviews come from candidates who use MeetAssist, so they lean towards software, data and AI roles: the shares describe these interviews, not every RAG interview. Answers to the most asked questions are written and checked by an editor; a "common mistake" is shown only when reviews of real answers in these interviews show it, with the count.
191 interviews, 122 candidates, Nov 2025 to Nov 2026. Updated every month with new interviews.
Or add the Chrome extension