Playwright

Playwright is our default for new web automation. Auto-waiting removes the category of timing bug that dominates Selenium maintenance, so suites we build on it tend to stay green without a standing babysitting cost. We lean on browser contexts to give every test a clean session cheaply, which means tests can run in any order and in parallel without the shared-state failures that force teams into serial execution. Traces are switched on for CI failures, so a broken run comes with a replayable recording instead of a screenshot and a guess. We use the built-in test runner rather than bolting Playwright onto an existing framework, and API request fixtures let us set up state directly instead of driving the UI through a login flow on every test. Where cross-browser sign-off matters, the same suite runs against WebKit and Firefox without a separate codebase.