How Rescale built an AI triage agent with Strands Agents and AI-DLC

Rescale, a digital engineering platform built for the AI era, faced the challenge of managing the operational overhead of its growing platform. Engineers were spending significant time classifying automated end-to-end test failures, which were critical for the timely deployment of services to production. To address this, Rescale developed an intelligent triage agent using the AI-Driven Development Lifecycle (AI-DLC) framework on AWS.

The AI-DLC methodology emphasizes a structured approach where AI executes tasks while humans make the crucial decisions. In the case of Rescale's triage agent, AI classified test failures into root cause categories and generated a remediation report with specific recommendations. However, the human engineers reviewed and validated each step before any action was taken. This ensured that the AI understood the team's intent and that nothing was left to assumptions.

During a five-day in-person workshop, AWS architects guided the development process while Rescale engineers worked together in real-time to build the agent. The team began with a vision document and technical environment document to capture the intent and ground the work in a technical environment. This approach helped them define clear requirements before starting the implementation.

The triage agent integrates seamlessly with Rescale's existing Argo Workflows CI/CD pipeline on Amazon Elastic Kubernetes Service (Amazon EKS). When a nightly Playwright suite of hundreds of test cases produces failures, the agent is triggered to classify each failure and generate a structured remediation report. The agent uses Claude Opus 4.8 on Amazon Bedrock for classification, retrieves relevant historical failure patterns from a knowledge base, and queries pod logs through Model Context Protocol (MCP) for runtime evidence. This comprehensive approach ensures accurate classification and efficient remediation.

By automating the classification and routing of test failures, Rescale's triage agent significantly reduces the manual effort required from engineers, allowing them to focus on higher-level tasks like architecture and product judgment. The agent has been successfully deployed in production within weeks, demonstrating the power of the AI-DLC approach in delivering rapid, production-ready solutions.