mabl is an AI-driven test automation company, and this weekly summary reviews notable product and technology developments that underscore its focus on AI-native quality engineering. The company continues to position its platform as a way for software teams to increase delivery speed while preserving strong verification and quality controls.
This week, mabl promoted an upcoming live customer session that will walk through a full “agentic testing” workflow using its platform, emphasizing a practical, slide-free format. The event will demonstrate how teams can integrate mabl into daily QA processes, including writing agent instructions that enforce testing standards automatically and building tests from real user journeys via its test authoring agent.
The session will also highlight conversational analysis of nightly test results, allowing users to query outcomes instead of relying solely on dashboards, and will focus on identifying coverage gaps before they cause failures. Live Q&A with the company’s AI experts is planned, reinforcing mabl’s effort to deepen engagement with existing customers and show practical use cases for its AI capabilities.
Recent engineering metrics previously shared by mabl showed merged pull requests rising from 179 per month in August to 810 in April, even as commit volume tripled. Despite this increase in activity, median time from commit to production fell from 24 hours to under 20 hours, suggesting that mabl’s internal delivery pipeline has become more efficient.
The company links these gains to a scalable verification layer built on two independent testing stacks and a production heartbeat that mirrors real customer paths. It has also highlighted AI-assisted peer review and an MCP server that feeds deployment test results back into its coding assistant, tightening the feedback loop between development and testing.
mabl has been migrating its MCP server to the cloud while enabling local test execution initiated directly from an AI coding agent, which caters to developers who favor terminal-based workflows. This approach keeps orchestration in the cloud but allows local runs, helping unify agent-driven development and testing and supporting more flexible engineering practices.
The firm also showcased an end-to-end MCP workflow for deployment monitoring, including a walkthrough by co-founder Dan Belcher that demonstrates how agents can inspect the latest deployment and cluster failures in real time. This workflow helps distinguish genuine regressions from tests that require updates, enabling faster triage and reducing manual QA overhead for DevOps and QA teams.
These updates collectively reinforce mabl’s positioning in the AI-enabled test automation and continuous delivery market by emphasizing real-time failure clustering, automated triage, and conversational interfaces. The focus on measurable efficiency gains and AI-driven workflows may enhance the company’s value proposition and support deeper adoption among enterprise customers as they seek faster, higher-quality software releases.
Overall, it was a constructive week for mabl, marked by continued investment in agentic testing experiences and verification-centric architecture that align with evolving demand for intelligent, automation-first quality assurance solutions.


