#rag

Articles tagged with #rag

Building a Hybrid Search Engine from Scratch: Combining BM25, Vector Search, and RRF
When building search systems, developers traditionally face a dilemma: Should we rely on exact keyword matching (like traditional databases or Lucene/BM25) or semantic context understanding (like mode
Aug 11, 20262 min read6
Deep-Dive into LLM Plumbing: Tokenization, Normalization, and RAG Architecture Mechanics
Building production-grade AI systems and Retrieval-Augmented Generation (RAG) pipelines looks glamorous from the outside, but true engineering lies deep in the invisible plumbing. If you've ever built
Jul 31, 20264 min read2