Hacker News Reader

Comment by NickC25 | Hacker News Reader

It most certainly IS possible.

That's the whole point of resumes and LinkedIn...especially in technical fields. You self-select for the hiring criteria you're looking for. You then speak to the candidate to understand a bit more about their experience as it relates to the job you're seeking a candidate for. That's why it often makes sense for such roles, for technical people to interview other technical people.

"Oh, you have experience with XYZ technology stack, and you worked with it for a few years, awesome. That's what we use here, but could you please talk about some of the projects you worked on with said technology?"

But no, tell me how that's not possible to do quickly.

View on HN

People lie about what they personally did as part of a larger team. They claim to have skills and expertise they don't actually have.

I'm not saying you need a gauntlet of tests. I liked the approach mentioned elsewhere of a technical discussion about what they did. But it takes, like, an hour or two to make sure they're not BSing and that they're actually as much of an expert as they claim to be.

Does that matter?

I mean, of course it would be better to find the most qualified candidates every time with perfect accuracy. Does having a longer interview process help with that goal, or does it deter the exact candidates you want? Is it okay to end up with the guy who wrote the FrobNizzle compactor instead of the woman who wrote the FrobNizzle extractor?

Great is God's Zoo.

The candidate may embellish and fuzz thing (both on the CV and verbally), mixing what they did with what the team did, color things to make themselves look better, etc. In certain parts of the world, that's just the normal behaviour in interview loops, and the loop takes it into account. Sometimes you remove the extra "marketing layer" and still have a good candidate --- sometimes not.

If you're Google or similar, you can afford to throw such candidates out even at the recruiter screening stage. But in other companies, probably even in the 3rd quartile, you may need to fish in these waters.

Example: does "shipped" mean it went into production and users were using it, or, does it mean you gave the source code to the client and then you have no clue what happened to it? Does "Airflow" on the CV mean you once played with it for a blog post, or you used it in production for 5 years? Does "Manager" mean you were managing the people and accountable for delivery, or you were in a lot of meetings?

You can weed out _some_ of the poor candidates in 15 minutes, but not _all_ of them, not even close. But the point of the interview loop is to weed out _most_ of poor candidates.

The fact that a somebody once got a job after a 15 minute interview is completely irrelevant.

Resumes can easily be bsed and 15 min can also be bsed using buzzwords. I know a person IRL who basically switches jobs every 3 months, but is excellent at using tech buzzwords, while not actually being able to code or think algorithmically, they know all the frontend tooling. More than usual frontend eng would know. So they appear stronger than an actual frontend eng. They can talk about webpack config and plugins, but they can't build anything.

They can make a thing that would take a day of work sound like some sort of fantastical world changing undertaking.

Aside from the fact that you'll eliminate a lot of good people, how does that let you screen out people who can't do anything but are decent bullshitters? In a 10-15 minute chat you might catch them contradicting themselves or say something you're lucky enough to remember is wrong (if you're lucky enough to know one of the things they claim experience with in detail), but that's not going to be reliable (and if you eliminate people who get product trivia wrong, that will also have a pretty high false positive rate).

I don’t think there’s really that many people who can successfully bullshit about being experienced and skilled technologists.

There’s something lacking with your interviewing skills if you can’t detect them in a conversation.

If on the off chance you do get some sociopath that somehow escapes all efforts of detection, well, it’s easy to fire people, in the US anyway.

> But no, tell me how that's not possible to do quickly.

I mean your low ball easy to answer question gives you exactly zero information other then yes they in fact worked on a team that did stuff. They just talk about what their peers did you've got such a short interview your not going to figure out they didn't actually do that shit.

Reply on news.ycombinator.com