How Rescale built an AI triage agent with Strands Agents and AI-DLC
Rescale used the AI-Driven Development Lifecycle (AI-DLC) framework on AWS, with Strands Agents and Amazon Bedrock, to build an intelligent test-triage agent in weeks. Learn how AI-DLC structures AI-assisted development and how Rescale is scaling the practice across its engineering organization.
First seen by Oriel on .
AI summary
Rescale used the AI-DLC methodology, Strands SDK, and AWS services to build a production agent that classifies nightly test failures, reducing time blocking deployments by 20%.
- The agent runs in Rescale’s Argo Workflows CI/CD pipeline on Amazon EKS, using Claude Opus 4.8 on Amazon Bedrock, historical failure patterns, and MCP-based log retrieval.
- It generates structured reports with root-cause classifications, file references, and recommended fixes for engineers to review; it does not automatically change code.
- A five-day workshop took the system from requirements to staging, with human review guiding scope and architecture decisions, including hosting on EKS and switching log retrieval to Amazon CloudWatch.
- Classification agreement with human triage increased from 44% at initial deployment to 68%, using iteration on a golden dataset and prompt engineering.
- A PyTest evaluation harness supports live and stubbed modes, while the team is extending AI-DLC to other engineering and simulation work.
#ai agents #ai-dlc #amazon eks #software testing
Generated by gpt-5.6-sol. Check the original for details.
Aliases
| Source | External ID | URL |
|---|---|---|
| AWS Machine Learning Blog | 7b42b92d3553785669b6b7f2bc67061098c8e566 | https://aws.amazon.com/blogs/machine-learning/how-rescale-built-an-ai-triage-agent-with-strands-agents-and-ai-dlc/ Read discussion |