Cheeses, vectors and PostgreSQL
summary
A demo uses a cheese.com dataset (id, name, origin, texture, description) and adds a vector column of embeddings (dimension chosen at model time) created with Hugging Face / OpenAI / Vertex. It stores embeddings in PostgreSQL with the pgvector extension and runs nearest-neighbor vector queries to return the three cheeses most similar to Saint-Nectaire; for comparison, a keyword LIKE filter returns 17 hits while vector similarity yields a ranked top‑3 that match real-life resemblance.
AI-written summary · about
related talks
50mA fun and absurd introduction to Vector Databases
52mpgvector Newbie to Expert
36mUnleashing the Power of Vector Databases for AI-Driven Applications
25mSemantic search with Django, PostgreSQL, & pgvector
48mFrom Text to Meaning: pgvector Transforms PostgreSQL Search
27mpgvector for Python developers
27mMachine Learning for Advanced AI Search
48mVectors are the new JSON