Database
pgvector development
How does AonHive use pgvector?
pgvector development for vector similarity search inside PostgreSQL.
pgvector is how we search by meaning without a second database. The embedding lives on the same Postgres row as the record it describes. A filter for tenant, status, or owner is a SQL predicate, not a step after the vector search.Discuss this stackHow we use it
The embedding sits beside the row.
- 01
Filter in the query
The search includes the access rule. A result the person cannot read is not fetched and then thrown away in the application.
- 02
Index for the shape you query
The index matches the distance function and the filter. We measure the query the product runs, not a benchmark on random vectors.
- 03
Keep the source text
The chunk that was embedded stays in the table. A hit can be shown and audited. The vector alone is not an answer.
pgvector
Bring the system this has to fit.
Tell us the product, the estate you already run, and what your team needs to own. We will say whether this belongs in the stack.