Questions
24 of 46
1What is RAG and why is it preferred over fine-tuning for domain-specific knowledge in production applications?
2What are the core components of a RAG pipeline in LangChain — Document Loaders, Text Splitters, Embeddings, Vector Stores, Retrievers, and Chains?
3What is the difference between semantic search and keyword search and why does RAG rely on semantic similarity?
4What is an Embedding in the context of RAG — what does it represent and why is cosine similarity used to compare them?
5What is a Vector Store and how does it differ from a traditional relational or document database?
6What is the difference between a Retriever and a Vector Store in LangChain — why is the abstraction separation important?
7What is a Document object in LangChain — what are pageContent and metadata fields and why does metadata matter in RAG?
8What are Document Loaders in LangChain and how do you choose the right loader for PDFs, web pages, Notion, Google Drive, or SQL databases?
9What is the difference between RecursiveCharacterTextSplitter and CharacterTextSplitter — when would you use one over the other?
10What is chunk size and chunk overlap in text splitting — how do you decide the right values for your use case?
11How do you handle structured documents like tables, code blocks, or markdown files during the splitting phase to avoid breaking semantic meaning?
12How do you load and split documents lazily (streaming) to handle very large files without running out of memory?
13What is a SemanticChunker and how does it differ from fixed-size character-based splitting?
14How do you preserve and propagate source metadata (filename, page number, URL, timestamp) through the loading and splitting pipeline?
15How do you choose the right embedding model — what tradeoffs exist between OpenAI embeddings, Cohere, HuggingFace, and local models like nomic-embed?
16What is the difference between dense embeddings and sparse embeddings (BM25) — when would you combine both in a hybrid search?
17How do you handle embedding model upgrades in production — what happens to your existing vectors when you switch models?
18How do you efficiently batch embed a large corpus of documents without hitting rate limits or memory constraints?
19What are the tradeoffs between vector stores like Pinecone, Weaviate, Chroma, pgvector, and FAISS — how do you choose for production?
20How do you implement namespace or tenant isolation in a vector store for a multi-tenant RAG application?
21How do you handle incremental updates to a vector store — adding, updating, and deleting documents without full re-indexing?
22What is HNSW indexing and why does it make approximate nearest neighbor search fast at scale?
23What is a similarity score threshold in retrieval and how do you use it to filter out low-confidence results?
24What is MMR (Maximal Marginal Relevance) retrieval and how does it balance relevance with diversity of results?
25What is a MultiQueryRetriever and how does it improve recall by generating multiple phrasings of the same question?
26What is Contextual Compression in LangChain retrieval and how does it reduce noise in retrieved chunks?
27What is a ParentDocumentRetriever — how does it index small chunks but return larger parent chunks to the LLM?
28What is HyDE (Hypothetical Document Embedding) and how does it improve retrieval for vague or abstract queries?
29What is Self-Query Retrieval and how does it allow the LLM to generate structured metadata filters alongside the semantic query?
30How do you implement hybrid search combining dense vector search with BM25 keyword search using EnsembleRetriever?
31What is a Re-ranker (cross-encoder) and where does it fit in the RAG pipeline after initial retrieval?
32What is the difference between Stuff, MapReduce, Refine, and MapRerank document chain strategies — when do you use each?
33How do you build a Conversational RAG chain that maintains chat history and reformulates follow-up questions into standalone queries?
34What is query decomposition and how do you break a complex multi-part question into sub-queries for better retrieval?
35How do you implement Step-Back Prompting in a RAG pipeline to improve retrieval for highly specific questions?
36What is CRAG (Corrective RAG) and how does it add a grading step to decide whether retrieved docs are relevant before answering?
37What is Self-RAG and how does the LLM decide when to retrieve, whether retrieved docs are relevant, and whether the answer is grounded?
38How do you implement a fallback strategy when retrieval returns no relevant documents — how do you avoid hallucination in this case?
39How do you implement RAG evaluation — what metrics like faithfulness, answer relevancy, and context recall do you measure using RAGAS?
40How do you detect and mitigate hallucination in RAG outputs — what role does citation and source grounding play?
41How do you build a citation system that maps each sentence in the LLM's answer back to the exact source chunk it came from?
42How do you handle multilingual RAG — embedding and retrieving documents in multiple languages for a global user base?
43How do you optimize retrieval latency in production — what caching, pre-fetching, or index optimization strategies do you apply?
44How do you implement access control at the retrieval layer — ensuring users only retrieve documents they are authorized to see?
45How do you handle long context RAG — when retrieved chunks exceed the LLM's context window, what strategies do you apply?
46How do you design a RAG pipeline with LangGraph — turning retrieval, grading, and generation into discrete stateful graph nodes?
langchainlangchain
Basics
Agents
Models
Messages
Tools
Middleware
Memory
MCP
Multi Agents
RAG
Context Engineering
Human in the loop
24 / 46

What is MMR (Maximal Marginal Relevance) retrieval and how does it balance relevance with diversity of results?

Maximal Marginal Relevance (MMR) balances query relevance and diversity among retrieved documents, reducing redundancy in search results.

Standard similarity search often returns very similar or duplicate documents. MMR addresses this by selecting documents that are both relevant to the query and dissimilar from each other. It uses a lambda_mult parameter (0 to 1) to trade off relevance (closer to 1) versus diversity (closer to 0).

MMR Retrieval with LangChain
Key MMR Concepts
  1. 1

    fetch_k: Number of documents initially retrieved; larger pools improve MMR’s selection ability.

  2. 2

    lambda_mult: Controls relevance-diversity trade-off (0 = max diversity, 1 = max relevance).

  3. 3

    When to use MMR: For summarization, QA over redundant datasets, or generating varied options.

Difficulty: 5/10
Topics: MMR algorithm, relevance‑diversity tradeoff, LangChain retrieval integration

Scenario Questions

0-2 years experience
  1. 1

    Suppose you have a LangChain Retriever that returns the top‑k documents based on cosine similarity. How would you modify the pipeline to use Maximal Marginal Relevance so the results are both relevant and diverse?

  2. 2

    If you set the λ parameter of MMR to 0.0 versus 1.0, what effect would you see in the returned documents?

  3. 3

    What happens if duplicate documents are fed into an MMR‑based retriever? How does the algorithm handle that?

2-5 years experience
  1. 1

    You notice that after enabling MMR in your question‑answering bot, the answers sometimes miss key facts that were present in the top‑k similarity results. Walk me through how you would investigate and fix the issue.

  2. 2

    When scaling the retriever to 10,000 documents, the latency of the MMR step becomes a bottleneck. What strategies could you employ to keep latency low while preserving diversity?

  3. 3

    Explain why changing the similarity metric from dot‑product to Euclidean distance altered the diversity of the MMR output.

5-8 years experience
  1. 1

    Design a retrieval pipeline for a multi‑tenant SaaS product that uses LangChain and must guarantee per‑tenant relevance and diversity while staying under a 200 ms SLA. How would you incorporate MMR, and what caching or indexing choices would you make?

  2. 2

    Your production system logs occasional spikes where MMR returns nearly identical documents despite a high λ value. What edge cases in the algorithm could cause this, and how would you redesign the component to be more robust?

  3. 3

    Discuss the trade‑offs between computing MMR on‑the‑fly versus pre‑computing a diversified candidate set for each query.

8+ years experience
  1. 1

    The company plans to migrate from a single‑node LangChain retriever to a distributed vector store (e.g., Milvus) while preserving MMR‑based diversification. What architectural changes are required, and how would you ensure consistency of the λ parameter across shards?

  2. 2

    How would you evaluate the long‑term maintainability of an MMR implementation that is tightly coupled to a specific embedding model, and what strategy would you propose to decouple them for future model upgrades?

  3. 3

    If multiple teams need different relevance‑diversity balances, how would you expose a configuration surface that allows per‑service λ tuning without breaking existing contracts?

Follow-up Questions

  • ❓Why would you choose a particular λ value for a given use case?
  • ❓What metrics would you monitor to verify that MMR is improving the user experience?
  • ❓How would you test the correctness of your MMR implementation?