As an ex-Dropboxer, Dropbox asks legitimately tough questions. I only got in because I got asked the exact set of questions that I could figure out the answer for, once I joined and went through interview training I realized I would have failed about half of the questions that Dropboxers ask.
I'm also not sure how it is at other companies (at Google but haven't gone through interview training yet), but Dropbox's rubrics are also pretty strict. Doing "well" on a question requires getting through multiple parts normally with close to zero bugs that you don't catch yourself.
You just described the entirety of the tech hiring experience. Sure some people are bonafide geniuses that can crack the hardest problems in their sleep. The majority of tech workers, however, are ones who got lucky with the specific combination of questions asked in the interview. Maybe they had seen the question before. Maybe the solution just "clicked". This is why the best strategy for acing such interviews is – just apply to a lot of companies.
Of course, but this is a gradient. Struggling startup may ask a question that 80% of engineers are capable of answering (through luck, exposure, experience -- whatever dimensions).
Dropbox asks difficult questions, and it's hard to discern why. I don't believe the problems at Dropbox are particularly difficult relative to its peers. I don't think they innovate at a clip that's outsized, etc.. But they do this, and their engineering culture focuses on this.
The joke internally is that Dropbox asks lots of concurrency questions because the Dropbox client has 50+ threads :-)
That said, I think what I noticed at Dropbox is that asking lots of tough questions gets you a lot of pretty talented folks who are very interested in solving hard technical problems. So from an infrastructure side, Dropbox was overflowing with talented people. From the product side, though, it was harder for teams to staff frontend projects or make progress when their ideas were challenged by infra.
It's easy to criticize the current state of tech interviews but I've never really heard anyone propose a better solution.
Startups ask questions 80% of engineers can answer because they don't have many applicants. Dropbox might have 50 decent applicants who could all probably do the job for every opening. How do you decide who gets it? Ask an easier question and you end up with way more passes than you have openings.
To be clear: not criticizing the current state of hiring. I agree nothing better exists, and it's a narrow mindset to think companies do this when there are better approaches.
My point is Dropbox asks questions that are beyond what its peers ask, even companies not its peer (FAANG, which we don't think Dropbox is apart of).
This is an explicit decision by the engineering leaders at Dropbox. It's interesting, and I'm wondering if it works for them.
I'm also not sure how it is at other companies (at Google but haven't gone through interview training yet), but Dropbox's rubrics are also pretty strict. Doing "well" on a question requires getting through multiple parts normally with close to zero bugs that you don't catch yourself.