Building a CI harness with Mastra

This workshop covers building a CI optimization harness with Mastra, using a research agent to propose code changes, validate them through A/B tests, and open pull requests backed by evidence. As coding agents generate more code, this approach helps teams tackle growing CI bottlenecks and distinguish real speedups from noisy benchmark results. Hosted by: Brandon Barros, Founding Product Advocate @ Mastra Daniel Worku, Co-founder & CTO @ StarSling CHAPTERS 00:00 Why coding agents are making CI a bottleneck 03:39 Designing the CI optimization loop 08:38 Running optimization experiments with Hatchet 12:38 Using ClickHouse to track experiments and give agents focused feedback 18:34 Validating speedups with A/B tests and confidence intervals 20:06 From agent proposals to verified PRs on production runners 25:41 Enforcing CI best practices with the CI score skill 30:32 Inside the Mastra research agent: prompts, tools, sandboxes, and traces 43:14 Results: faster CI for Better Auth and Mastra 44:40 Q&A: regressions, caching, experiment history, and agent progress WHAT IS MASTRA Mastra is an open-source TypeScript framework designed for building and shipping AI-powered applications and agents with minimal friction. It supports the full lifecycle of agent development, from prototype to production. 📚 MASTRA RESOURCES Website: https://mastra.ai X: https://x.com/mastra_ai Discord: https://mastra.ai/community/discord GitHub: https://github.com/mastra-ai Learn Mastra: https://mastra.ai/learn Principles of Building AI Agents (Book): https://mastra.ai/books/principles-of... Patterns for Building AI Agents (Book): https://mastra.ai/books/patterns-of-b...