Georgia State research center automates incident root-cause analysis from minutes to seconds using Amazon Bedrock and Strands Agents SDK
TReNDS Center at Georgia State University built an agentic pipeline that cuts manual root-cause investigation from 15–30 minutes to under 60 seconds by combining Amazon Bedrock, Strands Agents SDK, CloudWatch, Lambda, and GitHub integration.
1 source · cross-referenced
- TReNDS Center at Georgia State University built an agentic AI pipeline on Amazon Bedrock and the open-source Strands Agents SDK to automate real-time production error investigation.
- The system reduces manual root-cause analysis time from 15–30 minutes to under 60 seconds by enriching errors with log context and source code from GitHub.
- The pipeline uses Amazon CloudWatch subscription filters, AWS Lambda, Strands Agents SDK, and Amazon Bedrock to detect errors and deliver AI-powered analysis.
- TReNDS has run infrastructure on AWS since 2019, using Amazon EKS and CloudWatch with FluentBit for logs.
A research center at Georgia State University, TReNDS, built and deployed an agentic AI pipeline on Amazon Bedrock and the open-source Strands Agents SDK to automate real-time investigation of production errors.
The system reduces manual root-cause analysis time from between 15 and 30 minutes to under 60 seconds by automatically detecting errors, enriching them with log context, and pulling source code from GitHub for AI-powered analysis.
The pipeline integrates Amazon CloudWatch subscription filters to monitor logs, AWS Lambda for orchestration, the Strands Agents SDK for agentic workflows, and Amazon Bedrock to drive the AI analysis.
TReNDS has operated its infrastructure on AWS since 2019, running applications on Amazon Elastic Kubernetes Service (Amazon EKS) and shipping logs to Amazon CloudWatch using FluentBit.
The center, a joint effort of Georgia State University, Georgia Institute of Technology, and Emory University, develops advanced analytical methods and neuroinformatics tools for brain health research.
The architecture is described as running in production at TReNDS and is intended to automate the most time-consuming part of incident response: the root-cause investigation itself.
- Aug 8, 2026 · AWS — Machine Learning Blog
Cohere Health builds multi-tenant agentic architecture on Amazon Bedrock AgentCore to digitize clinical policies
Trust79 - Aug 8, 2026 · TechCrunch — AI
Airbnb reports faster feature development and tests AI-powered search toggle
Trust79 - Aug 8, 2026 · TechCrunch — AI
Rippling unveils AI Spend Console to track and cap employee AI tool spending
Trust79