# From problem to fix in minutes: Automating search relevance tuning with the Amazon OpenSearch Search Relevance Agent

DevFeed: [From problem to fix in minutes: Automating search relevance tuning with the Amazon OpenSearch Search Relevance Agent](<https://devfeed.tech/articles/from-problem-to-fix-in-minutes-automating-search-relevance-tuning-with-the-amazon-opensearch-search-relevance-agent-96232.md>)

Original publisher: [Read original article](<https://aws.amazon.com/blogs/big-data/from-problem-to-fix-in-minutes-automating-search-relevance-tuning-with-the-amazon-opensearch-search-relevance-agent/>)

Author: Aruna Govindaraju

Published: 2026-10-22T17:43:30Z

Content type: article

Language: en

Sources: [AWS Big Data Blog](<https://devfeed.tech/sources/aws-big-data-blog.md>)

Topics: [Amazon OpenSearch Service](<https://devfeed.tech/topics/amazon-opensearch-service.md>), [opensearch](<https://devfeed.tech/topics/opensearch.md>), [AI Chat](<https://devfeed.tech/topics/ai-chat.md>), [human in the loop agent](<https://devfeed.tech/topics/human-in-the-loop-agent.md>), [hybrid-search](<https://devfeed.tech/topics/hybrid-search.md>)

Tags: [2](<https://devfeed.tech/tags/2.md>), [agent](<https://devfeed.tech/tags/agent.md>), [amazon](<https://devfeed.tech/tags/amazon.md>), [amazon-opensearch-service](<https://devfeed.tech/tags/amazon-opensearch-service.md>), [complete](<https://devfeed.tech/tags/complete.md>), [data](<https://devfeed.tech/tags/data.md>), [example](<https://devfeed.tech/tags/example.md>), [experiment](<https://devfeed.tech/tags/experiment.md>), [intermediate-200](<https://devfeed.tech/tags/intermediate-200.md>), [investigation](<https://devfeed.tech/tags/investigation.md>), [language](<https://devfeed.tech/tags/language.md>), [minutes](<https://devfeed.tech/tags/minutes.md>), [offline](<https://devfeed.tech/tags/offline.md>), [opensearch](<https://devfeed.tech/tags/opensearch.md>), [post](<https://devfeed.tech/tags/post.md>), [query](<https://devfeed.tech/tags/query.md>), [relevance](<https://devfeed.tech/tags/relevance.md>), [search](<https://devfeed.tech/tags/search.md>), [service](<https://devfeed.tech/tags/service.md>), [technical-how-to](<https://devfeed.tech/tags/technical-how-to.md>), [this](<https://devfeed.tech/tags/this.md>), [with](<https://devfeed.tech/tags/with.md>)

## AI overview

The Amazon OpenSearch Service Search Relevance Agent automates a human-reviewed tuning loop: it analyzes User Behavior Insights (UBI) data, proposes query DSL changes, and validates them with offline Search Relevance Workbench (SRW) experiments before deployment. Its three skills handle behavior analysis, hypothesis generation, and evaluation; deterministic tools calculate metrics, and proposals expose their reasoning for review. In the example, the agent identified low overall clicks, investigated the "wireless earbuds" query (261 searches, 65 clicks, 24% CTR), proposed fixes for keyword field dilution, missing popularity signals, and `best_fields` vulnerability, then reported improved NDCG with minor MAP and Precision tradeoffs. The article says the agent does not alter queries at runtime; it tunes configurations for future searches. It requires an OpenSearch Service domain running OpenSearch 3.5 or later, the OpenSearch UI application, and behavioral data. A separate Hybrid Search Optimizer is described for tuning hybrid-search parameters.

## Source excerpt

The Amazon OpenSearch Service Search Relevance Agent automates search relevance tuning. You describe a problem in plain language, and the agent investigates your behavioral data, proposes an improved query, and validates it with an offline experiment. This post walks through the agent's architecture and a complete example investigation.