Mutation testing
Mutation testing measures test quality by introducing small bugs (mutations) into the source code and checking whether tests catch them. If a test suite has 80% coverage but kills only 40% of mutants, most of those covered lines aren't actually being tested for behaviour.
Tools like Stryker (JavaScript) and PIT (Java) automate mutation testing in CI. The cost is real — mutation runs are 10-100x slower than unit tests — so teams typically scope mutation to specific high-value modules (payments, auth, core data) rather than the entire codebase. The metric you care about: mutation score = killed mutants / total mutants. Anything above 70% is good; above 85% is excellent.
Long-form posts that explore mutation testing in depth — when to use it, common failure modes, how AI helps.