Vector Databases for AI Engineers
An Independent Guide to Indexing, Hybrid Search, Filtering, Scaling and Choosing the Right Vector Store
-
- $17.99
-
- $17.99
Publisher Description
Vector databases sit under semantic search, recommendations and retrieval-augmented generation, and the choice of store shapes the recall, latency, cost and safety of every AI feature built on top of it. Vector Databases for AI Engineers is an independent, vendor-neutral guide to how these systems work and how to run them well. It explains distance metrics and exact search, then the index families that make approximate search fast: IVF, HNSW, scalar, binary and product quantization, and disk-based graphs such as DiskANN, each with worked numbers. It covers filtered search and the pre-filter and post-filter trade-off, hybrid search with BM25, sparse vectors and reciprocal rank fusion, and multi-tenancy patterns. It then opens up storage engines, durability, updates and deletes, sharding and replication, and serverless designs that separate compute from storage. It surveys the landscape fairly across purpose-built databases, libraries such as FAISS, extensions such as pgvector, search engines and cloud services, and teaches honest benchmarking, cost modeling, security, data residency, operations, migration and a decision framework for choosing a store. One running example, a fictional online marketplace's semantic search team, ties every chapter together.
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.