This tutorial walks you through building a production-grade RAG pipeline using Claude as the generator. We’ll cover chunking, embeddings, reranking and citation extraction.
Step 1: Chunk your documents
from langchain.text_splitter import RecursiveCharacterTextSplitter
splitter = RecursiveCharacterTextSplitter(chunk_size=800, chunk_overlap=120)
chunks = splitter.split_documents(docs)Step 2: Embed and store
Use a small, fast embedding model. Voyage AI’s `voyage-3-lite` is currently the best price/performance balance.
Step 3: Generate with citations
Claude has first-class citation support — just pass your retrieved chunks as documents and it will return inline citations automatically.