I learned so much from solving leetcode problems that made dayjob coding easier. And my coding decisions are superior because I know all the different approaches and which is best for the exact situation I’m in.
Tbh mostly feels like grinding leetcode to change a job is a pain. People may just want to float their resume around and discuss experience.
People are lazy and don't wanna have to work hard even for fang pay.
Because it deflates the ego of the incapable. I actually love that there is a standardized set of problems to test engineers for a job at these big companies. It's much more fair that way. All you have to do is study, rather than having already had experience on these massive distributed systems
I used to think like this as well when I was earlier in my career, that LC interviews would level the playing field if I didn't have much exp. But not that I'm in my 30's I realize it's a bit of a chore to have to grind LC every time I want to switch jobs. At this point I'd like if I can just talk about my experience and knowledge rather than solve a bunch of timed algo questions :/
Leet code can be cognitively demanding and laborious at same time. This puts stress on people when they attempt it and that stress makes a lot of them quit. But they still want to believe they are as good as the guy who is putting time and effort. So they belittle the value of leet code to keep believing the lie.
I used to think it was useless and then I studied it and have been performing better at my job, able to distill problems into individual cases, point out better logical thinking in problem solving. I think the coding concepts are not directly applicable to the every day work, which is why it gets all the hate. But often LC is one of the faster ways to get signal from an interview.
It’s not so bad if you don’t have to do dynamic programming, which is just so fucking dumb. But LC mediums aren’t so bad.
Instead of do, just do a simple recursion, and add the Python decorator lru_cache, and say it's memoization. That gets you through like more than half of DP problems.
The problem is they aren’t good interview questions as they don’t signal much of value, and are prone to encouraging interviewers to be elitist dicks
I'm glad leetcode exists. Coding interviews are so hackable, yet people who can do them are still so low in supply that it keeps our TC up.
Yep. It's a cheat code to easy money that many people pass on. I know people who are getting half what they are worth because "they are too busy" or "don't like to interview."
I don't mind them as long as they don't ask the ones that require like a trick to be able to solve.
When you’ve solved 500 problems that require tricks, you now have 500 tricks in your pocket before the job. Compare that to figuring out each trick on the job and needing to sleep on the problem to see the trick the next day. Which engineer is moving faster and which is learning on the job at the company’s expense?
As an interviewee, it’s the ego check they don’t like. It makes them feel inferior. I personally think it’s great for shaking the rust off. As an interviewer, I don’t like the arms race it creates thanks to company questions leaking. If a candidate does too well on a difficult question where they just write the answer in silence for 10 mins, I get no signal on how they problem solve. For all I know they have the solution in front of them or have memorized all the company-tagged question solutions.
Yeah I agree that the arms race is the bad part. No one should ask hards.
ppl are lazy to spend time on LC including me, hence the hate