Recommendations
The Recommendations service delivers personalized video recommendations using vector similarity search. It ingests viewing signals and uses NVIDIA embeddings stored in Astra DB to find semantically related content.
Endpoints
| Method | Endpoint | Description |
|---|---|---|
| GET | /api/v1/users/{user_id}/recommendations |
Get personalized recommendations for a user |
| POST | /api/v1/recommendations/ingest |
Ingest viewing signal for recommendation model |
| GET | /api/v1/videos/{video_id}/related |
Get videos related to a given video |