Retrieval-Augmented Generation in Practice
$0.99
Buy on Amazon View on Apple BooksPrefer it in print? See every format on Amazon →
Price on Apple Books. Amazon’s price may differ.
- Published
- Genres
- Computers & Internet
Description
Retrieval-augmented generation (RAG) lets a language model answer from your own documents, with citations, instead of from what it happened to learn in training. Building a RAG demo takes an afternoon. Building one that stays accurate, fast, affordable and secure for thousands of users is an engineering discipline. This book teaches that discipline end to end, following one realistic system: an internal knowledge assistant for a global manufacturer, answering field engineers, support agents and employees from scanned manuals, regional policies, support tickets and a parts database in nine languages. You will learn how to parse difficult documents, choose chunking strategies, select and compress embeddings, size vector indexes such as HNSW and IVF-PQ, combine BM25 and dense search with reciprocal rank fusion, rewrite and route queries, rerank candidates, and generate grounded answers with verifiable citations that say "I don't know" when the evidence is missing. Advanced chapters cover knowledge graphs, GraphRAG, text-to-SQL, agentic, conversational and multimodal RAG. Production chapters cover evaluation with golden sets, recall@k, nDCG, faithfulness and model judges; access control, privacy, data poisoning and prompt injection; and freshness, latency, cost and observability.
This book is an independent publication and is not authorized, sponsored or endorsed by any company whose products it mentions. All trademarks belong to their respective owners and are used only for identification. This book was drafted with generative AI tools and checked with automated tools. Tools, features and prices change often; confirm them in current documentation.