Hunting down random test failures can be very very difficult, sometimes impossible, but minitest-bisect makes it easy. minitest-bisect helps you isolate and debug random test failures. If your tests only fail randomly, you can reproduce the error consistently by using `--seed `, but what then? How do you figure out which combination of tests out of hundreds are responsible for the failure? You know which test is failing, but what others are causing it to fail or were helping it succeed in a different order? That's what minitest-bisect does best.