Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Interviews in the Age of AI: Ditch Leetcode – Try Code Reviews Instead

chrlschn.dev

Post reply on HN
Print view View original

51–60 of 286 posts

Re: Interviews in the Age of AI: Ditch Leetcode – Try Code Reviews Instead

#51

Post by codeTired »

nfriedly wrote: post #46
codeTired wrote: post #36

Earlier quoted context omitted.

The problem is that it might have been true at some point but now there are resources to “learn” for interviews, so currently you will get people willing to grind. Nobody happy with their pay and job is going to grind leetcode. They could just ignore leetcode and ask puzzle problems instead to test ability.

Honestly, being "willing to grind" might be a good thing to target if it's a good proxy for being willing to do the work . However, I agree with the general sentiment that almost nobody is doing the research to ensure their interview process selects for the things they actually want.

I’m willing to grind if you present me with a real life problem you need solving. I’m not willing to run on a hamster wheel, it seems illogical to me.

Re: Interviews in the Age of AI: Ditch Leetcode – Try Code Reviews Instead

#52

Post by anthonypasq »

hcks wrote: post #30
bluehorizon2 wrote: post #26

People are forgetting why companies these days do Leetcode in the first place. The initial reason why we have Leetcode today is that Google originally determined through their interview research that the smartest candidates were the ones who were best at algorithms. Google wanted to hire the smartest people, not necessarily the best coders, so that’s why their interviews were mostly algorithmic. Of course everyone st…

It’s just a proxy measure for IQ really. They could have handed out Ravens matrices and got better outcomes prediction.

I assure you, they would if they could, but its illegal.

Re: Interviews in the Age of AI: Ditch Leetcode – Try Code Reviews Instead

#53

Post by m0llusk »

The big deal here is that it allows for relevance. Skilled coders don't want to waste time with exercises. They come up with useful tools and side projects and contribute to open source. Leetcode is a distraction because it is relevant only as an exercise, not as something truly useful. And that is the biggest test of a coder. Not that they can code something, but that they can apply the ability to code to real problems and address something relevant. Code that really matters has traction. Sometimes it is a very simple thing that makes a tool or application or library truly useful, and the skill to see that and apply such knowledge goes far above some leetcode exercise that has no real application and serves only for points in some game.

Re: Interviews in the Age of AI: Ditch Leetcode – Try Code Reviews Instead

#54

Post by Dudester230602 »

shortrounddev2 wrote: post #32
Aurornis wrote: post #29

Earlier quoted context omitted.

> runtime complexity has literally never once come up in my life There are definitely different types of programming jobs, but for some contrast: I have never had a job where runtime complexity did not come up as a significant factor somewhat frequently. It really comes down to what you’re doing, but at a certain scale or size of problem you really do need to know about algorithms, data structures, and runtime comple…

In my experience the biggest bottlenecks are in the network. Most optimizations are also very very low hanging fruit because the original code was written by a moron

Also, using scripting languages and misusing browser engines unironically yet expecting performance...

Re: Interviews in the Age of AI: Ditch Leetcode – Try Code Reviews Instead

#55

Post by bluGill »

osigurdson wrote: post #23
bluGill wrote: post #13

Earlier quoted context omitted.

There is research into how an interview should be done to find who will be a good or bad employee. Most companies (not just our industry!) don't appear to have even looked at it. (I'm not sure how useful the research is for real world hiring)

Links?

I don't know how to search the academic literature to find it. I'm not even sure it is online to link. I just know it is done.

Re: Interviews in the Age of AI: Ditch Leetcode – Try Code Reviews Instead

#56

Post by a13o »

I think a company with a mature code review culture wouldn't be able to benefit much from this kind of interview.

The candidate wouldn't have to figure out what the code does, because there would be a PR description and/or walkthrough video explaining the code.

There wouldn't be any bugs to find, because code reviews aren't a great way to find bugs. The bugs are found by automated and manual tests.

There would be no stylistic feedback to give. The linter already enforced everything enforceable, leaving only things that are a waste of time to fight about.

There's no point in giving architectural feedback, as the architecture was already socialized and agreed upon before the code was written.

Now on to what code reviews are good for. The changeset is probably a small piece of a bigger picture, because the code is being shipped incrementally. The candidate has no comprehension of the bigger picture to weigh in on downstream issues.

The candidate has no broader understanding of the codebase in general, to suggest reusing an existing pattern, or avoiding a common pitfall with a private API.

The candidate can be well-versed in the OSS framework, and suggest better ways to write some particular boilerplate, or avoid a framework pitfall.

I fear an interview structured like this says more about the company than the candidate. If your company's development lifecycle is mature, there's not much a candidate can weigh in on.

The most useful case that jumps out to me is gaining an understanding of the candidate's familiarity with a framework. A contrived diff falling into lots of common pitfalls could cover more ground than having them write something in the framework. But having them write something in the framework shows you a lot more than just their awareness of pitfalls, so I'm not convinced this is better still.

Re: Interviews in the Age of AI: Ditch Leetcode – Try Code Reviews Instead

#57

Post by wiremine »

Dudester230602 wrote: post #40
wiremine wrote: post #33

I'm a CTO for a small software studio. We've used mini projects and code reviews in our interview process for years, and it works great. The code review is a great way to explore a candidate's ability to code, and also to communicate their decisions about a piece of code. In the "age of AI", I'm not sure we'll ditch the mini projects. I think testing a candidate's ability to _use_ AI to generate code is an important…

Would you allow employees to subcontract and outsource their work?

> Would you allow employees to subcontract and outsource their work?

No, but I do encourage them to use the best tools possible. :-)

Re: Interviews in the Age of AI: Ditch Leetcode – Try Code Reviews Instead

#58

Post by galangalalgol »

anthonypasq wrote: post #52
hcks wrote: post #30

Earlier quoted context omitted.

It’s just a proxy measure for IQ really. They could have handed out Ravens matrices and got better outcomes prediction.

I assure you, they would if they could, but its illegal.

A quick search indicates that it is not illegal in the US as a whole, but maybe California?

Re: Interviews in the Age of AI: Ditch Leetcode – Try Code Reviews Instead

#59

Post by Dudester230602 »

wiremine wrote: post #57
Dudester230602 wrote: post #40

Earlier quoted context omitted.

Would you allow employees to subcontract and outsource their work?

> Would you allow employees to subcontract and outsource their work? No, but I do encourage them to use the best tools possible. :-)

"Sir, there is this approach where I feed in requirements and get back results, can I use it?"

Re: Interviews in the Age of AI: Ditch Leetcode – Try Code Reviews Instead

#60

Post by wiremine »

Dudester230602 wrote: post #40
wiremine wrote: post #33

I'm a CTO for a small software studio. We've used mini projects and code reviews in our interview process for years, and it works great. The code review is a great way to explore a candidate's ability to code, and also to communicate their decisions about a piece of code. In the "age of AI", I'm not sure we'll ditch the mini projects. I think testing a candidate's ability to _use_ AI to generate code is an important…

Would you allow employees to subcontract and outsource their work?

Just to add to my previous comment. We use AI heavily today, and we're seeing a 25%+ productivity improvement. And our Devs are happier, as they get to focus on high order problems.
Post reply on HN