RAG (Retrieval-Augmented Generation) is a technique that improves AI responses by retrieving relevant external knowledge before generating an answer. Embeddings convert text into vectors so that semantic similarity can be computed between queries and documents. Chunking is the process of splitting large documents into smaller pieces so retrieval becomes more accurate. A vector database stores embeddings and allows fast similarity search using nearest neighbor algorithms. In production systems, retrieval quality is more important than the model itself.