Vector Search — how HNSW finds nearest neighbours

Originally published on software-engineer-blog.com.

Your documentation site has 2,000,000 help articles. A user types a question into the search box and gets the best match in 2 milliseconds—without the system ever comparing that question to almost any of them. This is not magic. It's HNSW: Hierarchical Navigable Small World, the graph structure that powers vector search in FAISS, pgvector, Qdrant, Weaviate, and Milvus.

Mental model: HNSW is a multi-layer graph where every vector points to its nearest neighbors, and search is a greedy walk that descends from sparse long-range links at the top to dense fine-grained links at the bottom—like taking highways down to local streets to find an address.

The Problem: Why Brute Force Is Correct but Unusable

You already have the pieces:

Originally published on software-engineer-blog.com.

The Problem: Why Brute Force Is Correct but Unusable

You already have the pieces:

Vector Search — how HNSW finds nearest neighbours

Vector Search — how HNSW finds nearest neighbours

Related reading

How Vector Search Actually Works: IVF and HNSW

Building a Vector Search Engine from Scratch: The Math and Mechanics of HNSW

I Built a Vector Search Engine from Scratch — Here's What I Learned

Vector Search at Scale: Why Your Index Isn't as Healthy as You Think

Advantages of Building a Vector Search Solution

Vector Strike: Semantic Search Database Defender

Related reading

How Vector Search Actually Works: IVF and HNSW

Building a Vector Search Engine from Scratch: The Math and Mechanics of HNSW

I Built a Vector Search Engine from Scratch — Here's What I Learned

Vector Search at Scale: Why Your Index Isn't as Healthy as You Think

Advantages of Building a Vector Search Solution

Vector Strike: Semantic Search Database Defender