Advanced RAG
$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
You have built a retrieval-augmented generation (RAG) system. It works on the demo questions, then fails on the hard ones: multi-step questions, tables in scanned PDFs, questions about a whole corpus, users who must not see each other's documents, and an index that is always a week out of date. This book is the next step for engineers who already know the basics. Following one realistic system, the research knowledge assistant of a fictional global pharmaceutical company, it goes deep on query understanding and query planning, late chunking and hierarchical indexes, hybrid search with learned sparse retrieval, cross-encoder rerankers and ColBERT-style late interaction, agentic and iterative retrieval, self-correcting designs such as Self-RAG and Corrective RAG, GraphRAG and knowledge graphs, and multimodal retrieval over page images, tables and figures. Production chapters cover long context versus retrieval with worked cost numbers, caching at every layer, personalization, permission-aware and multi-tenant retrieval, freshness and incremental indexing, evaluation with offline sets, calibrated model judges and RAGAS-style metrics, observability, cost and latency engineering, and security against data poisoning and prompt injection through documents.
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.