News for hackers. Stuff that matters.
Re:Ask HN: What to do about ‘Good at programming Bad at Leet... by beej71 on Friday May 20, 2022 @08:21PM (#31454145)
> I’ve spent the last year or so reinventing the hiring process at my tech company to expressly avoid algorithmic puzzles and leetcode nonsense, and I was able to do this because I patently refused to accept the status quo.

I think companies that break out of the leetcode habit are going to do well. There's a lot of great talent out there that's being passed over by these tests.

If the job is to solve leetcode problems, by all means, test the candidates for that. If the job is to solve real life problems, there's almost certainly a better measure.

I will never judge someone based on that kind of problem. I believe the data that results is just bad.

Edit: I should add that leetcode is useful for developing programming and problem-solving skill. I've done tons of problems, myself, and have fun solving them... and learn from them. So I do recommend it for a challenge. But not for interviews.

Re:Ask HN: What to do about ‘Good at programming Bad at Leet... by stutsmansoft on Friday May 20, 2022 @09:06PM (#31454360)
> I think companies that break out of the leetcode habit are going to do well. > There's a lot of great talent out there that's being passed over by these tests.

Agreed. And there's a lot of great talent that doesn't even want to try with some of these companies solely due to the hostile interview process.

Re:Ask HN: What to do about ‘Good at programming Bad at Leet... by kabdib on Saturday May 21, 2022 @12:15AM (#31455189)
I recently took a year off, and started interviewing in March. I spent a fair amount of time in fear of interviewing, doing LeetCode pushups and feeling miserable about my performance, and wondering if I could swing retirement rather than go through another fucking hazing ritual. Excuse my language.

I resolved up front not to submit to any kind of take-home test, proctoring software, or algorithmic nonsense unrelated to the actual job of shipping software. A polite "No thank you" was what I planned. But the plan was thwarted!

I was pleasantly surprised by the technical interviews I actually got. They were mostly nicely constrained problems that were real-world and not algorithmically nasty. One was a little unintended "collaboration" in the form of helping a senior dev debug their own "what is wrong with this code?" exercise (it probably went smoother for the next candidate they interviewed). In short, it was actually kind of fun.

I started a new dream job this week. At 61 this is no longer easy, but the company is enlightened enough to support me while I get up to speed, and it's fun to help mentor the folks just getting into the industry.

There are decent companies out there. Hopefully the terrible ones will start getting a clue.

Re:Ask HN: What to do about ‘Good at programming Bad at Leet... by tsuujin on Friday May 20, 2022 @08:25PM (#31454168)
Strong agree with all of this.

Ideally, interview questions should be directly related to the domain of the position. Show the candidate what they will actually be doing if they accept an offer.

The whole process should be a two way conversation.

Re:Ask HN: What to do about ‘Good at programming Bad at Leet... by bfung on Friday May 20, 2022 @10:20PM (#31454725)
Agreed. As a hiring manager, I’ve also explicitly, purposely come up with coding questions that reflect normal work: call APIs & issues, sync data issues, model data for usable systems.

The hires, then, focus on the right customer problems and don’t spin on creating random, useless code.

Everyone from mgmt to ICs, technical and non-technical, are happier.