Muthu Pitchaimani

Your AI agents and assistants are only as good as the answers they retrieve. Out-of-the-box frameworks get you started, but production-ready AI applications demand more from their search layer including higher relevance on complex queries, persistent memory across sessions, and tools the model can invoke reliably.

Learn how to build next-generation observability that goes beyond traditional search and analytics. In this session we will discuss how organizations can reduce operational costs by modernizing their observability stack using OpenTelemetry, OpenSearch, and S3.

This chalk talk delves into the art of right-sizing Amazon OpenSearch Service clusters for optimal cost and performance. It explores the trade-offs between provisioned clusters and Amazon OpenSearch Serverless, tailored to different workloads such as log analytics, lexical, semantic, and sparse search.

Enterprises have been building logging solutions to gain visibility into their operations for well over a decade. In this session, learn how Amazon OpenSearch Ingestion helps you ingest operational data such as logs, traces, and metrics to aid your observability needs.

Search engines use text analysis to power your search. In this workshop, learn how machine learning (ML) transforms traditional search techniques with ML techniques for ranking, relevance, content, and query understanding.

With OpenSearch and OpenSearch Dashboards, you can store, analyze, visualize, and monitor your application, infrastructure, and security. There are many choices and tradeoffs to consider as you ingest your data and work with it in OpenSearch Dashboards.

Amazon OpenSearch Service’s Trace Analytics functionality allows you to go beyond simple monitoring to understand not just what events are happening, but why they are happening. In this workshop, learn how to instrument, collect, and analyze metrics, traces, and log data all the way from user front ends to service backends and everything in between.

Amazon Elasticsearch Service (Amazon ES) supports a wide range of search use cases, including e-commerce, application search, web search, data lakes, and more. Manually searching through a vast amount of data to get accurate and relevant results is challenging.