First Orion Enhances QA Automation with Amazon Nova Act

First Orion, a provider of branded communication solutions, faced challenges in maintaining QA efficiency amid rapid product development and a growing portfolio of web applications. Their traditional automation frameworks relied heavily on fragile, selector-based scripts that frequently broke with UI changes, causing delays and increased manual effort.

To address these issues, First Orion adopted Amazon Nova Act, an agentic AI platform that allows QA teams to describe tests in plain English. This shift enabled their QA analysts to author and execute tests directly, without dependence on automation engineers or brittle selectors, significantly reducing test maintenance and execution times.

The implementation involved building a comprehensive architecture where QA test cases are authored via a React-based UI, stored in Amazon S3, and executed through a Python-based test runner on Amazon ECS, leveraging Amazon Nova Act SDK and AgentCore Browser for autonomous browser interactions. The system supports dynamic data generation, parallel execution, and detailed reporting, integrating with Allure dashboards and Microsoft Teams for notifications.

Since deployment, First Orion has observed a 20-25% reduction in QA cycle times for certain testing scenarios and estimates a 25-30% saving in engineering capacity. The early detection of regressions and increased test coverage have contributed to higher product quality and faster release cycles.

Key lessons from their experience include the importance of organizational support, flexible planning, and clear prompt engineering for AI models. They also highlight the need for explicit context in test case definitions, as AI agents do not share institutional knowledge.

Looking ahead, First Orion plans to expand automated test generation using AI and code change insights, aiming for a fully automated testing pipeline that further enhances efficiency and product quality.

Overall, their adoption of Amazon Nova Act exemplifies how agentic AI can transform QA workflows, enabling faster, more reliable testing processes that scale with product growth.