Flaky Test
A test that passes and fails intermittently on the same code, undermining trust in the whole suite.
What is Flaky Test?
What is Flaky Test?
A flaky test is one that yields different results on identical code, usually because it depends on timing, ordering, shared state, or unstable selectors. Flakiness is corrosive: once a suite cries wolf, developers start ignoring red builds, and real defects slip through. The cures are determinism-first design—proper waiting instead of sleeps, isolated state, and resilient locators. QI tackles flakiness at the source by generating Playwright code that auto-waits and by healing selectors when the UI shifts, so failures mean something.