Leetcode made perfect sense very early on, before leetcode, the company, existed. It was, at the same time, part a test of smarts, having some capacity to work under pressure, whether a new grad had paid any attention to the data structures and algorithms classes, while we got to see that they could code at all. It was an improvement over whiteboard interviews. And thus, it was what separated a green college student from very high compensation at generous tech companies. The ideal candidate wouldn't even prep for this: they'd just show up, remember the right things, let their talent shine a little, and go past. The questions weren't even that hard back then: You might not even touch what now is a leetcode medium in a typical interview loop. It came from the DSA classes because it would be easy to explain, and something you might recall and implement in 30-50 minutes. It couldn't all just be fizzbuzz.

The problem is that the compensation was so generous, and the kind of questions being asked so public, that your typical somewhat less talented, yet hard working grads would just cram for the interviews, and get past them, thus lowering the value as a filter. Leetcode and the like decided that there was an industry in helping you cram. The filtering got worse and worse, and the companies then decided to keep upping demands, hoping to still get the most talented. But there's nothing in a leetcode interview that cannot be trained. So the metric became the target, and we ended up in the current mess.