Skip to content
Quantoros

What are the key components of a RAG pipeline?

LLM & AILevel: FoundationCore Concepts

What are the key components of a RAG pipeline?

Hints
  • Break the query into candidate retrieval, context selection and generation.
  • How would you locate the stage that lost a fact without using only one end-to-end metric?

Answer

ComponentOptionsKey Decisions
Document loaderLangChain loaders, PyMuPDF, UnstructuredFile formats, OCR for scanned docs
ChunkerFixed-size, recursive, semanticChunk size, overlap strategy
Embedding modelOpenAI ada-002, E5, BGE, CohereDomain fit, dimensionality, cost
Vector storePinecone, Weaviate, Chroma, pgvectorScale, metadata filtering needs
RetrieverDense, sparse, hybridRecall vs precision tradeoff
Re-rankerCohere rerank, BGE-reranker, cross-encoderQuality vs latency
LLMGPT-4, Claude, LlamaQuality, cost, latency
OrchestrationLangChain, LlamaIndexDeveloper experience

Source and licence: JustInCache & contributors — Awesome AI Interviews · Apache-2.0

Related questions