Part 4 — Vector Search Foundations

This section builds the retrieval substrate underneath agent memory: embeddings, indexes, hybrid search, chunking, and the precision/recall trade-offs that decide what gets recalled.

What you will learn

  • How embeddings, ANN indexes, and distance metrics work in practice
  • When keyword and hybrid search beat pure vector retrieval
  • Chunking, reranking, filtering, and freshness for memory workloads
  • How embedding drift and query formulation affect recall quality

Chapters in this section

There are 24 chapters in this part. Open any chapter to read it on its own, or work through them in order.

  1. 1 What Is a Vector Embedding?
  2. 2 Embedding Models and Semantic Similarity
  3. 3 Vector Indexes: How Approximate Nearest Neighbor Search Works
  4. 4 HNSW: The Graph-Based Index Behind Modern Vector Search
  5. 5 Distance Metrics: Cosine, Dot Product, and Euclidean
  6. 6 Why Vector Search Alone Is Not Enough for Memory
  7. 7 Keyword Search and the Inverted Index
  8. 8 BM25: Scoring Relevance Without Embeddings
  9. 9 Hybrid Search: Combining Vector and Keyword Retrieval
  10. 10 The Alpha Parameter: Tuning the Balance Between Semantic and Lexical Search
  11. 11 Reranking: A Second Pass for Precision
  12. 12 Late Interaction Retrieval Explained
  13. 13 Multi-Vector Representations
  14. 14 Vector Quantization and Compression Trade-offs
  15. 15 Filtered Vector Search: Combining Metadata and Similarity
  16. 16 Chunking Strategies and Why They Matter for Memory
  17. 17 Hierarchical Chunking for Long Documents
  18. 18 Late Chunking: Preserving Context During Segmentation
  19. 19 Recall vs Precision in Memory Retrieval
  20. 20 Approximate vs Exact Search Trade-offs at Scale
  21. 21 Index Rebuilding and Vector Freshness
  22. 22 Embedding Drift: When Meaning Shifts Over Time
  23. 23 Choosing an Embedding Model for a Memory System
  24. 24 Query Formulation for Effective Memory Recall