I've heard myself and other say many times, "the kind of work they have you do in the technical interview is rarely the kind of work you'll actually do". This so often true, I think many of us take it for granted. Odds are strong that you won't be implementing a hash table, a question I've been asked in several technical interviews. Often when pressed on this kind of thing, I'll say something like, well isn't it good to know anyway?
I think the author makes an important point when they say...
> My technical skills have definitely sharpened up by practicing Hackerrank-style questions, but I believe the companies using these tests are creating a bias in favor of candidates who a) have time to practice and b) who've recently graduated and are familiar with the format. It's very hard to prepare for these tests if you're working full time and have a young family. It's also an opportunity to excuse ageism. In an assessment the other day I was asked to code something that I hadn't covered for 30 years. If I'd had the opportunity to read-up on it, I would have been fine.
Again, I think it's taken for granted that FANNG companies should be gated in this way since they are looking for the "best of the best". Maybe it's worth taking a harder look at this and actually try to get an idea of who these "best of the best" look like. Are they all young people with wealthy families who attend top-tier computer science programs and apply for these jobs soon after graduation? I suspect that is the case.
Many of these hiring techniques do trickle down to other, smaller companies who view whatever Google or Facebook might be doing as a "best practice". I think it's worth asking if this kind of bias is acceptable.