From the source
Rescale built an AI triage agent with the AI-Driven Development Lifecycle framework on AWS, taking it from problem definition to staging deployment in five days and to production within weeks.
- The agent automates classification and routing of nightly end-to-end test failures, a task that had become a full-time job hidden inside an on-call rotation.
- It runs as a step in Rescale's existing Argo Workflows CI/CD pipeline on
Amazon EKS, triggered when a nightly Playwright suite of hundreds of test cases produces failures.
- The agent invokes Claude Opus 4.8 on Amazon Bedrock to sort each failure into one of four root cause categories: infrastructure issue, code bug, flaky test, or environment problem.
- It then pulls historical failure patterns from a knowledge base and queries pod logs through Model Context Protocol for runtime evidence.
- The agent completes the workflow in seconds and writes a structured remediation report with file references, function names, and recommended fixes for engineer review before any action is taken.
- The engagement was a five-day, in-person workshop where AWS architects guided the build while Rescale engineers worked hands-on-keyboard.
