Flake rate
Flake rate is the proportion of test runs that fail intermittently without any change to the code under test, and it is the primary measure of whether a test suite can be trusted.
Above roughly one percent, engineers begin re-running failures instead of reading them, and the suite stops functioning as a safety net. Flake is nearly always caused by timing assumptions, shared state between tests, or test data that is not isolated — and it is fixable, though usually by deleting or rewriting rather than by adding retries.
Related terms
- Self-healing tests
Self-healing tests use resilient element location strategies — and often a model to resolve ambiguity — so that an automated test continues to work when the underlying markup changes but the user-facing behaviour does not.
- QA agent
A QA agent is an LLM-driven tester that explores an application by reading its interface, deciding what to try next, and reporting what appears broken — covering exploratory testing that fixed scripts cannot.
Tell us what you are building
Send us the problem in a paragraph. You will get a straight answer on whether we can help, and what we would do first.
Book a call