Hacker News Reader

That may be true but what proponents of Leetcode style interviews say, is that they don't care. This process filters out good and bad candidates but "post filter", what remains is higher quality candidates. It is absolutely questionable from a scientific perspective (where is the tracking of the "B" group for instance), but if companies want to do irrational things they are allowed to do it.

View on HN

I think this wouldn't even be that bad of a thing (assuming that there's just so many candidates that you need some kind of filter), if it weren't for the issue pointed out in the study cited by the article: "But here’s a surprising finding: not a single woman in the public setting passed, while every woman in the private setting did."

That to me implies this isn't just an imperfect filter, but actually a very biased filter.

I expect for some companies, that pay 3-8X above average there should be a very large number of candidates. The really odd situation is the low paying, enterprise CRUD job that feels the need to use leetcode hard on their tiny pool of candidates that can't get jobs anywhere else.

Reply on news.ycombinator.com