How are you supposed to use LeetCode?

Think of the leetcode site as a set of practice problems. If you're struggling to solve a problem, then that shows what you should study to become better at leetcode. In your case, maybe the O(n) algorithm is a pattern that's broadly applicable to other problems too.

I think that if you're stuck on a problem for more than 30 minutes, it's better to go read the solution and learn from it than to keep struggling. (Then go back a week later and see if you can solve the problem from memory).

Opinions certainly differ on whether leetcode makes for a good interviewing process. On your specific problem, there are a ton of similar problems where you need to traverse possible solutions in a particular order to be efficient. I regularly see similar problems at work.

I don't think it's that non obvious. You're maximizing the area A = H*W which means maximizing H and W. Obviously the largest W is between height[0] and height[n-1]. You don't know if this will give the largest A as there might be a pair of heights with smaller W so you have to search inside. If you decrease W but want to increase A, then H must increase. From this you should recognize that the only way to increase H is with an inner line that's higher than min(height[0], height[n-1]). Any other lines cannot mathematically be candidates for a larger area. Once you know how to select your candidates, it's just a matter of checking them to find the max area.

Of course trying to make this deduction during an actual interview when one or multiple people are watching you is an entirely different problem. Hopefully leetcode will give you enough practice to recognize patterns so that you don't need to think too deeply during interviews.

> But it's the kind of thing you would see in a CS textbook, not just come up with on your own.

Well, LeetCode is a CS-textbook-problems kind of site.

You use it however you like. While I was searching for my first job, I spent some time solving problems and optimizing them, half to keep my skills from degrading after college, and half just for fun. I just treated it like a puzzle collection and had a good time with some of the more open-ended questions. I'm sure it'd be useful to help learn a new language as well. I'm not convinced it's effective as a general skill-improvement tool, though.

I think leetcode interview is useful to assess the candidate reasoning in front of a problem. If you don't produce a code that runs correctly in 30 mins, this should not matter as long as you have explained your approach and reasoned about options and complexity. Of course, not all company reason like this. I guess in FAANG interview they want you to nail the exercises during the interview. I think for them this is a signal of exceptional capabilities and a proof that you real worked hard to try to get a job with them.

It actually comes down to a few simple algorithms, only one of which I didn't see in college. These kinds of water problems use a "monotonic stack" if I remember correctly. That's the algorithm that I didn't see in school.

In reality, after you practice a lot of these problems, you realize that there are only a few broad categories and that most of them just use what you learned in school. You just have to practice to become fast.

I wish I could help here. I have made exactly one foray into leetcode, wrote out an algorithm for a fairly easy tree traversal thing, tried to submit it and just was told it went into an infinite loop.

I copied my solution elsewhere ran it, and it worked fine.

The only thing I could guess is the input wasn't what I was expecting, but I couldn't verify what input leetcode was actually giving me, so I just walked away.

The whole thing was a waste of time anyways. I still have yet to ever get a leetcode question during an interview and I've been doing this almost a decade now.

Then again, I don't try and get jobs at FAANG companies or FAANG wannabes.

You beat your head against the wall, submit your solution and see how it compares runtime-wise and memory-wise with others. If you're not scoring well on those metrics, then you go to the discussion to see if there are better solutions and re-implement and resubmit with the goal of understanding how the solution works. After doing this enough times you start to notice patterns of solutions to different classes of problems. If you want something that is more guided at a regular pace, have a look at interviewcake.com

It’s a spiritual digression. Our best developers didn’t come up from a hazing. We failed in assessing talent, so we resorted to this.

We have yet to see outsized results from the leetcode group, but we have seen outsized results from the pre-non-leetcode group.

Results matter, as the leetcode dipshits would say.

Do popular 300 medium difficulty problems

the idea is that you can produce the optimal solution within a minute in a tech interview. to do that you have to memorize it because no-one is able to do that otherwise. not even that gayle woman who wrote that shitty book. so, this is why coding interviews are broken today.