Vol. III · Issue 06 · May 2026Free worldwide PDF + EPUB delivery · No DRMISSN 2814-9921
PDF·EPUB·Lifetime updates
Machine Learning Systems · 1st Edition · January 2026
Vector Search At Scale
ANN indexes, recall budgets, and the database under your RAG
4.7(134 ratings)
intermediate
296 pages
RAG demos are easy. RAG that serves 10k QPS without melting your bill is engineering. This manual covers HNSW vs IVF vs DiskANN tradeoffs, recall budgets, hybrid search with BM25, sharding strategies that survive index rebuilds, and the failure modes specific to filtered ANN queries. Includes benchmarks across Qdrant, Weaviate, pgvector, and a from-scratch HNSW implementation.
Author
Dr. Priya Anand
ML Platform Engineer
Priya runs the kind of ML platforms where a 200ms regression costs more than your annual cloud bill. Her writing focuses on the boring infrastructure that makes models actually serve traffic.
$19.99
Instant PDF + EPUB delivery
DRM-free, copy onto any device
Free chapter updates for the life of the edition
Specifications
- Pages
- 296
- Edition
- 1st Edition
- Language
- English
- Level
- intermediate
- ISBN
- 978-1-99999-010-4
- Published
- January 2026
Editorial review
Reviewed by three working engineers at peer publications before publication. We do not publish first drafts.
Table of contents
What you'll find inside.
- 01The ANN Landscape
- 02HNSW From Scratch
- 03IVF and PQ Tradeoffs
- 04DiskANN for Bigger Indexes
- 05Recall vs Latency Curves
- 06Hybrid Search That Actually Helps
- 07Sharding and Rebuilds
- 08Filtered ANN: The Hard Problem
Also in this section
