How important is top-of-mind CS knowledge?
From my last two interviews, I feel like I need a part-time job just to drill CS puzzles, as I'm not flexing those muscles at all in my current fullstack dev role.
It depends on the role and the career level you're looking for, although some companies still do the algorithms questions, but it also depends on the interviewer, I like the case study approach because we can put you on a situation that is close to what will be your day to day job and also see how you implement it, so far it has had very good results and we are very happy with that.
Probably for other positions or companies the algorithmic challenges work best, I don't claim to have the secret sauce and we are always learning and improving our process and for us it works to find out if someone has what it takes through a conversation than while drilling them with complex and most of the times hypothetical problems.
Unless the job you're applying for is heavily focussed on developing algorithms to solve hard problems, technical interviews which focus on those type of problems should be considered red flags. In fact, any coding interview which cannot realistically tell the employer something useful about your ability to perform the tasks in the job description with the tools/languages required is a definite red flag, IMHO.
I am 100% agree with Kevin, any decent company should offer you a lot of transparency on the role and the responsibilities related to it.
Sadly, this is all too common, at least in Silicon Valley. Apparently, the tech giants have decided that puzzles and algorithm questions are the way to find the best engineers. I'd like to know whether they have metrics which prove that because over 10+ years of interviewing, I've found the opposite to be true.
So much of what we do now revolves around effective use of tooling and understanding how to simplify code and workflows. Particularly with front-end, there is virtually no need to devise clever algorithms. I've seen that people who look for clever solutions tend to overthink simple problems and write more (complex) code.
I want to know how you can quickly solve practical, real-world problems, so I started giving candidates a mini-app to build, and I've found that that process reveals more about them than giving them a series of algorithm/puzzle questions.
We're a place where coders share, stay up-to-date and grow their careers.
We strive for transparency and don't collect excess data.