[bin_bash]

> I was motivated and was able to solve most easy and medium question in 30 mins. However, that was not good enough for FAANGs.

What makes you think it's not good enough? LC is important and today it might be even be essential to land a role, but I think people tend to think of it like an exam when that's not really what we're testing for.

The goal is to see how you think, not see if you can simply come up with the right answer. If you need a hint it's usually not going to hurt you at all so long as you listen to the hint, are able to understand it, then incorporate it into a solution. I'd go so far as to say that being able to do that can help you more than reciting an answer without a hint.

There are some patterns like tries, left/right pointers, memoization, heaps, and maps that studying helps with for sure, but that doesn't mean you're supposed to be able to immediately solve something without any help.

(Note that there are asshole interviewers out there that will challenge you and won't follow what I've described, but that's rare and personally I've only seen them at startups, not big tech)

[8ytecoder]

All we need is one interviewer who absolutely hates interviewing to ruin the experience for the candidate and probably throw them off balance. It happens in Big Tech much more often than people are willing to admit.

I get where you are coming from when you say "the goal is to see how you think" - I really do and I have definitely believed in that once upon a time. Truth is, it's extremely gut/instinct driven than we realise. So yeah, while you can easily shoot down a candidate who seem to regurgitate what they had already solved, there's just absolutely no way to differentiate it from a candidate having a bad moment (let alone a bad day).

I had a take home exercise recently where I was supposed to identify a performance issue and fix it. I read the code in a hurry after finishing my day job. I knew where the likely issue is coming from but just couldn't locate it. I wrote back as such. The next day the solution came to me. I fixed it and sent it back. I still got hired but that sort of thing can never happen in a leet-code type interview no matter how much we'd like to believe. I have ADHD but even otherwise our brains are finicky.

[bin_bash]

My point is just that being able to solve LC mediums under 30 minutes isn’t that relevant. I’m not saying it’s a great way to vet candidates.

30 minutes is probably about what it takes me to solve an LC medium on average and I’ve passed every single FAANG full onsite loop I’ve taken which is a total of 6 over my career. I almost always need a hint or two if it’s not a trivial problem.

If I practiced I could get that number down but I don’t think it would make it more likely to get a better offer.

[granshaw]

Wait wait wait - have you seen the Amazon interview process? The first assessment step is a completely automated LC test with no human on the other side. Either you pass the all the test cases including your algo being quick enough on large datasets or you don’t.

Yes they are this blatant about it nowadays.

[bin_bash]

I’ve gone through the process twice with Amazon as recently as last month and not had this screen. I’m sure they offer it for some candidates but it’s not going to be LC medium difficulty.