Database
Qdrant development
How does AonHive use Qdrant?
Qdrant development for vector search with rich filtering.
Qdrant fits retrieval that needs payload filters, a predictable latency budget, and the option to run the store ourselves. Points carry the payload the query must match. The text they came from stays in the operational database.Discuss this stackHow we use it
Filter and search in one query.
- 01
Put the rule in the payload
The filter is part of the search, not a loop after it. A point without the tenant id is a bad write.
- 02
Size it for the query
Collection settings follow the dimension, the distance, and the filter the product uses. A default that was never measured is not a design.
- 03
Keep a rebuild path
The source text can refill the collection. The vector store is derived. Losing it does not lose the record.
Qdrant
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.