IDInternals Decodedininternals-decoded.hashnode.dev·3h ago · 11 min readThe ReAct Loop: How Every AI Agent Actually WorksThroughout the AI System Design series we follow Mailmind, a fictional AI email assistant that reads an inbox, triages messages, and drafts replies. It is our running example. Every AI agent that che00
IDInternals Decodedininternals-decoded.hashnode.dev·3d ago · 10 min readEvaluating RAG: Retrieval Metrics That Predict Answer QualityA RAG (retrieval-augmented generation) pipeline lives or dies on whether the retriever finds the right evidence. The metrics that actually predict answer quality are recall under your token budget, co00
IDInternals Decodedininternals-decoded.hashnode.dev·5d ago · 11 min readGrounding and Citations: Answers You Can CheckIn Part 5 we saw how hybrid search and re-ranking can dramatically improve retrieval quality. Now we tackle the next challenge: making sure the answers built from those retrieved chunks are actually t00
IDInternals Decodedininternals-decoded.hashnode.dev·Sep 28 · 13 min readHybrid Search and Re-Ranking: The Cheapest Quality WinHybrid search runs two retrieval methods in parallel, BM25 for exact keyword matching and vector search for semantic similarity, then fuses their results using reciprocal rank fusion. Adding a cross-e00
IDInternals Decodedininternals-decoded.hashnode.dev·Sep 26 · 12 min readVector Databases: What Actually Matters When ChoosingVector databases look remarkably similar on the outside. Most wrap the same open-source ANN libraries and offer search, insert, delete over HTTP. The parts that actually change your RAG (retrieval-aug00
