Stop Wasting Hours on Test Unpredictability and Get Back to Builds You Can Trust

Your test fails. You rerun it. It passes. You rerun it again. It fails somewhere else. That is not bad luck. That is evidence. In this full-day hands-on workshop, you will work through real flaky test examples and turn random-looking failures into failures you can reproduce, understand, fix, and prevent. The examples use JavaScript/TypeScript and Playwright, but the learnings apply to other programming languages and testing frameworks as well. You will find flaky tests with repeated runs, changed execution order, parallel execution, evidence from continuous integration, logs, screenshots, traces, and small experiments. Then you will trigger failures due to timing issues, shared test data, test-order dependencies, resource conflicts, differences in continuous integration, and external systems. You will inspect the evidence, decide whether the failure is flaky or real, and fix the cause yourself. You will also use agentic artificial intelligence tools to speed up the hardest parts of flaky-test work: reproducing intermittent failures, running targeted experiments, comparing results across repeated runs, and improving continuous integration checks to catch new flaky tests earlier. You will practice deciding when the tool is helping, when it is guessing, and when a suggested fix only hides the problem. Join me to stop losing afternoons to tests that fail but pass when you rerun them. You will leave with a practical way to catch flaky tests early, reproduce them reliably, fix the root cause, and trust your pipeline again.