Before adopting Claude AI for Playwright automation and building a Claude-powered QA Workspace(QA- Testbot), the client’s QA process involved significant manual effort across the end-to-end testing lifecycle. QA engineers had to manually analyze Jira tickets, understand requirements, identify test scenarios, create and maintain Playwright test scripts, execute tests, investigate failures, and prepare test reports. This made the overall QA cycle time-consuming and created dependency on experienced automation engineers for converting business requirements into executable tests.

The existing automation process also required considerable effort to maintain test scripts as application workflows, lenders, and requirements changed. Test execution and result analysis were largely separate activities, requiring engineers to switch between Jira, automation code, test environments, logs, and reporting tools. This fragmented workflow made it difficult to quickly identify failures, understand their root causes, and provide timely feedback to development and business teams.

As the volume and complexity of testing increased, the client needed a way to reduce repetitive QA activities, accelerate test creation and execution, improve automation coverage, and bring the complete QA lifecycle into a more streamlined workflow. The goal was not simply to introduce AI into testing, but to leverage Claude AI to create an end-to-end QA capability that could assist from Jira ticket analysis → test case generation → Playwright automation → test execution → failure analysis → reporting, with minimal manual intervention.

Trantor assisted FairSquare during the initial adoption of Claude and helped integrate its capabilities into the day-to-day QA lifecycle. The team matured the client-created Claude workspace plugin(QA-Testbot) capabilities through practical QA usage, applying them for Jira analysis, test generation, execution support, and reporting.

In parallel, Trantor initiated AI-powered Playwright automation, using Claude to accelerate framework and test-script development, execution and debugging, script maintenance, and reporting. This iterative adoption transformed Claude from an initial AI capability into a practical, mature, AI-assisted QA solution.

Claude-powered QA practices delivered significant improvements in speed and efficiency, with automation development becoming approximately 50% faster and end-to-end QA cycle time reduced by approximately 60%. For example, the team completed 43 SBL redesign tickets in just 4 days, compared with an estimated 10–11 days using the traditional manual approach, enabling faster testing, automation, maintenance, and reporting.