Products
What Relevan gives your agents.
Relevan is a database engine as a service, built for AI agents. You load your documents, products, profiles, or other proprietary data into an index you own and control. Your agents search that index, and Relevan learns from the searches they run.
Query, inspect, improve. We run the cluster.
- Explainable Search API
- Query endpoint returns ranked results plus explanation signals.
- Responses expose why each document matched, not just scores.
- Built for retrieval loops that need deterministic, inspectable behavior.
- Schema Exploration API
- Discover available fields, data types, and constraints at runtime.
- Let agents plan queries from actual schema state, not hardcoded assumptions.
- Reduce failed calls and token-heavy retries caused by missing field context.
- Feedback and Tuning API
- Submit interaction outcomes and quality annotations.
- Incrementally tune relevance without rebuilding your full search pipeline.
- Report a click, an agent judgment, or a conversion with one call to the feedback endpoint.
- Managed Search Infrastructure
- Relevan runs and operates the OpenSearch cluster behind your index.
- Your team is not responsible for tuning or scaling that infrastructure.
Two calls: search, then report the outcome.
Your agent searches the index. When a result leads to a click, an agent judgment, or a purchase, one more call tells Relevan.
RELEVAN_URL="https://fix-console-sidebar-brand.api.dev.relevan.dev"
# Search the index
curl -X POST "$RELEVAN_URL/v1/$ORG_ID/indexes/products/search" \
-H "x-api-key: $RELEVAN_API_KEY" \
-H "content-type: application/json" \
-d '{"query": "travel headphones"}'
# Report the outcome. Send back the queryId
# from the response's searchContext.
curl -X POST "$RELEVAN_URL/v1-beta/$ORG_ID/indexes/products/feedback" \
-H "x-api-key: $RELEVAN_API_KEY" \
-H "content-type: application/json" \
-d '{"actionName": "conversion", "label": "purchase",
"queryId": "<queryId>", "objectId": "sku-12345"}'Working with a coding agent?