Qdrant Feature
Qdrant feature
The qdrant feature adds a Qdrant vector database gRPC client wired into the composition root. Use it for semantic search, embedding storage, and vector similarity retrieval.
What it provides
| File | Purpose |
|---|---|
internal/adapters/persistence/qdrant/client.go |
Qdrant gRPC client connection |
Tech stack
| Library | Purpose | Documentation |
|---|---|---|
| qdrant/go-client | Official Qdrant Go gRPC client | qdrant.tech/documentation |
| Qdrant | Vector similarity search engine | qdrant.tech |
Learning resources
- Qdrant documentation — qdrant.tech/documentation (quickstart, collections, vectors, filtering)
- Qdrant Go client — github.com/qdrant/qdrant
- Qdrant recipes — qdrant.tech/documentation/examples
- Vector similarity search guide — qdrant.tech/articles
Notes
- Connection settings are configurable in
configs/config.yamlunder theqdrantsection. - Qdrant uses gRPC for communication — the config includes host, port, and API key fields.
- For local development, start Qdrant with Docker:
docker run -p 6333:6333 -p 6334:6334 qdrant/qdrant. - Combine with embedding models (e.g., from OpenAI, Cohere, or sentence-transformers) to store and search vectors.