Let’s break down the problem: * hiring is a very high-stakes decision * the normal interviewing process has no hope of directly measuring a candidate’s fitnes

Let’s break down the problem: * hiring is a very high-stakes decision * the normal interviewing process has no hope of directly measuring a candidate’s fitness for a particular position. There simply is not enough time. You’d want to evaluate the candidate over maybe a couple weeks of full time work, which, of course, a large number of candidates aren’t willing to do. * So you have to use the limited time you have to get as much information as possible and accept that direct knowledge really just cannot be obtained. The idea of leetcode exercises is that someone capable of the analytical problem solving skills and determination necessary to do well at those puzzles will also be capable enough to handle the harder parts of software development jobs. It’s an imperfect approximation, but for a problem where there is no closed-form solution, it could be one of the better of a number of poor options. (That’s the idea anyway, I don’t know if it’s true.) All that said, I think a code review is a good idea. Maybe do a couple leetcode puzzles and a code review (and a design discussion, and a “explain a recent project of yours to me” discussion, etc)