Embeddings turn any piece of text into a list of numbers (a vector) that captures its meaning. Two pieces of text with similar meanings get similar vectors — which is the whole trick behind semantic search, recommendations, and RAG (retrieval-augmented generation).
/The typical setup
- 01Add a vector column: Lovable adds a pgvector column to the table you want to search.
- 02Embed on insert: Every new row is embedded automatically via a trigger or edge function.
- 03Query by similarity: Take the user's search text, embed it, ask Postgres for the closest rows.
It just feels smarter
Users can search "how do I get my money back" and hit a doc titled "Refund policy". No keyword overlap required.