Skip to content

Embeddings & Vector Databases

Meaning as coordinates, search as distance.

embeddingssimilaritycosine similaritysemantic searchvector databasesPineconeWeaviateQdrantpgvectorANN indexes

Lessons

  1. 22.1Embeddings and vector databasesEmbeddings turn meaning into map coordinates, so "find something similar" becomes "find what's nearby". 30 min Embedding space
  2. 22.2Similarity metrics: cosine, dot product and Euclidean distanceThere are three common ways to say how close two vectors are. They can disagree, and knowing when saves you from slow, wrong searches. 30 min Embedding space
  3. 22.3Choosing a vector databaseEvery vector database does the same core search. What really decides the choice is everything around it, and often the answer is the database you already have. 30 min RAG builder

On the job: try it on your own project

Find a search or matching feature in your product, or a place where users struggle to find things. Ask Claude Code to explain how it works today and to prototype semantic search with embeddings on a small sample, on a branch. Compare ten real queries. Write a short decision note: keyword, semantic or both, and which vector store you would use and why.

Can you do this now?

0/4. Anything you can't tick yet: reopen that lesson's practice ladder.