Ask HN: Does LeetCode/HackerRank make you a better developer?

I prefer the adjectives strong/weak over good/bad or better/worse to describe developer skill for two reasons. Good or bad can be perceived as an inherent quality, while it is obvious that strength is acquired. The second reason is because better/worse can imply subjectivity. That is, that it wouldn't be possible to tell a strong developer from a weak one because it is all in the eyes of the beholder or some BS like that.

Onto your question, the answer is yes, very much yes. Solving puzzles indeed makes you stronger. It exercises your brain exactly like lifting weights exercises your muscles. Almost any skill can be trained the same way; there are chess puzzles, writing "puzzles", acting "puzzles", even musicians have certain "puzzles" they hone their techniques on (e.g Segovia scales).

I don't know why puzzles have such a bad rep on Hacker News. It may be a case of sour grapes. Like you see a jacked dude and you think "Eh, easy with steroids!" or "Yeah, but does he know basic arithmetic?" Cause admitting that he spent countless hours in the gym while you (like me) spent countless hours watching movies and eating chocolate is a too big of a blow to your confidence.

Apparently some dude didn't land a job with Google because he couldn't invert a binary tree. So he thought it was useless. I looked it up (https://leetcode.com/problems/invert-binary-tree/) and had a laugh. Because the solution was readily obvious and required no more than three lines of Python.

I don't necessarily think that practicing leetcode will make you a better developer, but good developers should be able to solve most leetcode questions in a few hours from scratch (with full internet access) without having ever seen those types of problems or data structures involved - maybe not with perfect computational efficiency, and certainly not in 30 mins during an interview

That said, writing more code and solving harder problems is what makes you a better developer - so if leetcode is what gets you to do that then it will probably have the desired effect

My exact experience, just few more years. And no, I’ve never heard from anyone in my (decent above average) pay grade about their production leetcode success story. Not even a glimpse of it, not a single time.

I myself though implemented BFS once for a factory which planned tight supply for scheduled orders. That was it. Highest CS I ever needed.

do you think that being able to solve LeetCode/HackerRank questions (not brute forcing them, but rather understanding the underlying algorithm) help you become a better developer

Being aware and understanding what it does is enough, like in “it does x in log” or “it does x in nlogn with these caveats”. I’d never try to reinvent/reimplement any of it and would object if anyone tried. Because it’s trivial to screw up an edge case, and given the amount of screw ups in plain dumb code… I don’t believe anyone who claims they could do it in the wild. It must be an academic setting, not a regular company at least.

I don't know, but they certainly killed my interests in CS.