Momentic has unveiled a major update to its software testing platform, adding autonomous features designed to help engineering teams keep pace with faster code delivery. The company also introduced a refreshed brand and said the platform is now open for any software team to try.
The announcement comes as software teams increasingly rely on AI coding tools, which can speed up development but also expand the volume of changes that need to be tested. Momentic argues that quality assurance has become a larger bottleneck as more code reaches production more quickly. In its release, the company cited recent industry reports suggesting incidents tied to AI-generated code are rising.
Momentic said the latest version of its platform is designed to behave less like a traditional test runner and more like an always-on QA system that understands how a product is supposed to work. The company described the update as the result of close work with AI-native engineering teams whose products serve large user bases across productivity, media, consumer, and professional services.
A central addition is the Knowledge Base, which lets teams encode product-specific definitions and expectations into the platform. According to Momentic, teams can document terminology, define intended behavior, and distinguish between bugs and acceptable changes. That shared context is then used by agents across the system, including those that write tests, sort through failures, or suggest fixes.
Momentic said the goal is to preserve the kind of institutional knowledge that experienced testers often build over time, but in a form that can be reused by software agents and new team members.
The company also introduced Explore Agent, a feature that monitors pull requests, reads code changes, and proposes new or revised tests based on what changed. Momentic said the feature is meant to reduce the manual work involved in keeping test coverage current as products evolve.
Instead of asking engineers to author test scripts from scratch for every update, the system is intended to identify affected user flows and extend coverage automatically. Momentic said the agent becomes more effective over time as it learns more about the product and the team’s testing preferences.
Another part of the update is a Failure Classification Agent, which analyzes test failures and sorts them into categories such as real bugs, intentional product changes, setup problems, or temporary errors. If a failure is caused by an expected application change, Momentic said the system can open a pull request to adjust the test. If it detects a likely bug, the platform surfaces an alert with contextual information for engineers.
Momentic also introduced a new test format that it says is easier for both humans and AI systems to understand. The company is positioning the format around plain-language intent rather than low-level script details, with the aim of making tests simpler to create, read, and modify.
Alongside the product changes, Momentic launched a new visual identity that it says reflects the platform's focus on structure, reliability, and autonomous behavior. The company said the refreshed design is meant to match its updated product direction.
Momentic said the platform is now available for any software engineering team and can be tried for free. The company framed the release as part of a broader effort to make quality assurance less manual and more deeply embedded in the development process.
The update positions Momentic among a growing number of companies trying to use AI to reduce the friction of software testing, particularly as development teams ship code faster and face more pressure to catch issues before they reach production.