Learn how to evaluate, track, and improve search relevance.
Connect to Elasticsearch, OpenSearch, Vespa, or any search system.
Group and manage queries to evaluate search performance.
Define reusable search request formats for each endpoint.
Run evaluations to measure ranking quality with metrics like NDCG and MRR.
Automate relevance grading with the LLM-as-a-judge pattern using OpenAI, Anthropic, Bedrock, and more.
Collect relevance grades to benchmark search quality.