AI killed the tech interview. Now what?

Post by codr7 » Fri, Feb 21, 2025, 3:21 AM UTC

ghaff wrote: post #466

codr7 wrote: post #461

Earlier quoted context omitted.

Employers trying to hire for fit and culture has resulted in the most inhumane and counterproductive processes I've witnessed. If that's what you really want to do, let the person work for at least a month in the team.

Provisional hires (which often exist in theory but they're pretty much a formality in general) don't work for the most part. Lot of overhead for the company. And in many cases you're asking for the candidate to quit a job and possibly relocate on the possibility they'll get a new position assuming that they click in a short time interval.

Post by edmundsauto » Fri, Feb 21, 2025, 3:47 AM UTC

rmah wrote: post #331

endemic wrote: post #186

Earlier quoted context omitted.

I think the argument is that for a good chunk of business work, you don't need to use the "interesting"/"deep" concepts. Sure, you'll need time to adapt to the idioms of the language you're using, but following examples you can be just as productive as others in a relatively short time.

But companies don't pay high salaries for the 80% mundane and easy tasks you do day to day. They pay for the 20% that is challenging and especially for that 1% of problems that occur only once every few months or years. If that 80% was 100% of the job then the company could pay 1/2 to 1/3rd the amount by outsourcing it.

Post by tptacek » Fri, Feb 21, 2025, 4:47 AM UTC

callingbull wrote: post #575

kasey_junk wrote: post #573

Earlier quoted context omitted.

Your citation addresses it. Less than 1% of employment lawsuits are about selection criteria and employers win over 90% of them. They suggest GMA tests are _more_ defensible than other approaches. The most interesting thing in that paper is that years of experience performed so poorly. It’s in the lowest cohort. Worse than “interests” or more general “biographical data”.

There is still the reputational risk of using selection methods with widely known disparate outcomes. Other methods also have disparate outcomes, but most of the criticism is directed at IQ tests. I've heard "IQ tests are culturally biased" but never "work sample tests are culturally biased", and I'll guess that's the experience of most hiring managers too.

Post by physicles » Fri, Feb 21, 2025, 4:55 AM UTC

leeny wrote: post #584

We did an experiment at interviewing.io a few months ago where we asked interviewees to try to cheat with AI, unbeknownst to their interviewers. In parallel, we asked interviewers to use one of 3 question types: verbatim LeetCode questions, slightly modified LeetCode questions, and completely custom questions. The full writeup is here: https://interviewing.io/blog/how-hard-is-it-to-cheat-with-ch... TL;DR: - Interview…

Post by Atotalnoob » Fri, Feb 21, 2025, 5:05 AM UTC

macintux wrote: post #513

eikenberry wrote: post #422

Earlier quoted context omitted.

IMO interviewing is the biggest bottleneck and if interviewing was decoupled from hiring then it wouldn't be a problem. But this requires a Guild-like organization to manage interviewing/vetting and for companies to use said Guild for hiring. The companies could then do a single team culture meeting (if they wanted) before hiring.

I wish I could remember who, but there is a company out there who conducts tech interviews to create a pool of candidates for their customers. Pretty sure there was a post here about it, but it's lost in my ocean of unread favorites.