San Francisco-based Momentic has announced a major platform update that introduces what it calls the first agentic quality platform, designed to help engineering teams verify software at the speed of AI-generated code. As AI coding tools enable teams to ship more code faster, traditional validation methods have struggled to keep up, leading to bugs reaching production and teams spending time on hotfixes instead of new features.
The new platform features several AI agents that operate on a shared product knowledge base. The Explore Agent automatically expands test coverage by identifying untested areas and generating tests. The Failure Classification Agent categorizes test failures and can automatically open pull requests to fix non-bug issues. The platform also introduces an intent-based test format that is readable by both humans and AI, making tests easier to maintain and understand.
In early beta results, Momentic's agents analyzed over 70,000 test failures, opened 400 pull requests, wrote 600 tests with a 73% merge rate, executed more than 2 billion steps, and verified over 80,000 pull requests. These figures demonstrate the platform's potential to significantly reduce the manual effort required for software quality assurance.
Companies such as Notion, Xero, Bilt, Webflow, and SPS Commerce are already using Momentic to prevent incidents from reaching customers as code volume scales with AI. The platform supports web, Android, and iOS applications, offering comprehensive cross-platform coverage.
Momentic was founded in late 2023 and is backed by Standard Capital, Y Combinator, FCVC, Transpose Platform, Dropbox Ventures, and Karman Ventures. The company is headquartered in San Francisco, California. For more information, visit momentic.ai.
The launch underscores a growing need for verification tools that can match the pace of AI-assisted development. As AI continues to scale code output, platforms like Momentic's could become essential for maintaining software quality and reducing engineering time lost to test maintenance.
