I think that a blanket rejection of the "Leetcode" style interview is too broad. They are a tool, and like any tool, one should try and understand when they ar

I think that a blanket rejection of the "Leetcode" style interview is too broad. They are a tool, and like any tool, one should try and understand when they are appropriate for the job. If the "the job" is to try and provide a fairly level playing field on which to assess candidates on raw problem-solving ability and programming intuition, these are useful tools. There are the same problems as with any test of this kind (you can improve your results by learning the format of the test well), but there's still a lot of signal. If you're trying to hire based on the things these tests measure, you should you use them. As the post points out, correctly, those are far from the only qualities required for many jobs, and anecdotally it seems like some companies are overusing this one type of interview, but the solution is for interviewers to assess what they are trying to select for, and make sure that their selection process measures it as well as possible. Part of that process may well still be puzzle-style interviews.