@admitsWrongIfProven Please don't delete. Questions are the best. I practice TDD where I drive all production code through a failung test. This means I write a failing test first. And only after that, I write the code to make it pass. This ends up me spending more time writing the test - going slow, thinking about the design. The writing of the test is the act of designing here. So in my workflow, there is no "write the tests fast". The tests are first class, and they come first. The production code is what I try to write fast to get to green quickly.

The fixing of a flaky test that fails once a year will pay at one point. Assume how long it will take you to fix it, and the cost of the failure, and do the math.