Course catalog
Search by meaning, not keywords.
Modern AI search runs on vectors: text, images, and code turned into points in high-dimensional space where nearness means similarity. Start with embeddings and the models that produce them, learn the distance metrics that rank results, then master the ANN indexes (HNSW, IVF, quantization) that make billion-scale search fast. Finish by building real semantic and hybrid search in Postgres with pgvector and choosing the right engine from Pinecone, Weaviate, Qdrant, Milvus, and Chroma.
Section 1
Turn text into numbers that carry meaning. Learn what an embedding is, how meaning becomes geometry, and which models to reach for in 2026.
Section 2
Once meaning is geometry, ranking results is just measuring distance. Learn cosine, dot product, and L2, how to pick one, and the curse of dimensionality.
Eligible first-time members can get 7 free days · Cancel anytime
Get startedSection 3
Make search fast at scale. Learn approximate nearest neighbor indexes (HNSW, IVF), quantization for compression, and the recall-versus-latency dial you control.
Section 4
Ship real semantic search in Postgres with pgvector, then survey the dedicated engines. Vector columns, distance operators, HNSW indexes, metadata filtering, hybrid search, and how to choose.