onelinkLearn RAG from zero: 10 links, in order
Retrieval-augmented generation: get an LLM to answer from your own documents. Read top to bottom; each link builds on the one before. Prompt to make your own: "I'm a developer new to RAG. Give me a learning path of 10 links, beginner to production, and save it to onelink."
What is RAG? (Pinecone) The clearest plain-English intro. Start here. https://www.pinecone.io/learn/retrieval-augmented-generation/
The original RAG paper (Lewis et al., 2020) Skim the abstract and figure 1. That's enough to know where the idea came from. https://arxiv.org/abs/2005.11401
Sentence Transformers Make embeddings locally for free. Good for understanding what an embedding actually is. https://sbert.net/
MTEB leaderboard Compare embedding models by task before you pick one. https://huggingface.co/spaces/mteb/leaderboard
Chroma docs The easiest vector database to start with. Runs in a notebook. https://docs.trychroma.com/
pgvector Already on Postgres? Add vectors there instead of running a new database. https://github.com/pgvector/pgvector
Build a RAG app (LangChain) First end-to-end build: load, split, embed, retrieve, answer. https://python.langchain.com/docs/tutorials/rag/
LlamaIndex The alternative framework, stronger on document loaders and indexes. https://docs.llamaindex.ai/en/stable/
Contextual Retrieval (Anthropic) The fix for "retrieval found the wrong chunk". Read it once your first version works. https://www.anthropic.com/news/contextual-retrieval
Ragas Measure whether answers are actually grounded. Do this before you tune anything. https://docs.ragas.io/