We all know this moment: when a bug we have never thought of before pops up right at the last moment in production. Truth is, forecasting all potential bugs with our tests is barely infeasible. Yet, what if there was still a way? Let's investigate together what can be improved and added to our test strategies to make them even more reliable thanks to property based testing. But contrary to classical talk around property based testing, we are going to go further by extending it to user interfaces and race conditions detection. By the end of this talk, our CI will have been turned into an automated QA identifying issues in our UI.