Hacker News Reader

Comment by Kranar | Hacker News Reader

It's because in many cases the years of experience doesn't count for much, that's why. The sad truth is I get plenty of resumes from people with 20+ years of experience and when I interview them to see how they actually write code, the code they write is not much different today than it was 20 years ago.

There's a difference between someone who has 20 years of genuine experience, and someone who has 3 years of experience that they've been repeating over and over again, and unfortunately my experience is most people who've been at this job for a long time are simply those who gained a few years of experience until they got to a point of diminishing returns and then allowed their skills to stagnate to the point where a recent grad could easily ramp up to out compete them with about 1 year worth of training and investment.

View on HN

> a recent grad could easily ramp up to out compete them with about 1 year worth of training and investment.

What about if you gave the veteran programmer 1 year worth of training and investment?

The whole point is that I'm mostly indifferent about whether someone is a veteran programmer with 20 years of experience or is a recent grad with 2-3 years of experience. The number of years of experience matters very little to me. Both get the same technical interview, get asked the same questions, and are evaluated on the same basis.

I think it's reasonable to wonder if the interview process is actually "indifferent", or whether it's designed to screen out older people in favor of younger people. How exactly could you or would you "test" for the value of experience, or do you believe that experience never has any value?

I test for competency not experience. If your experience makes you competent then great, if it doesn't make you competent then so be it, I don't care one way or another because I'm not looking to hire experienced people but rather competent people.

> I test for competency

Well, that's what everyone thinks they're testing for. The empirical evidence suggests otherwise. https://news.ncsu.edu/2020/07/tech-job-interviews-anxiety/

You again.

If what you're implying is that experience makes somebody anxious when performing under pressure, then obviously that's a rational reason to prefer those who have less experience.

I don't think experience means more anxiety though. And you don't know what the interview process of the GP is like. Why assume they are not telling the truth?

> You again.

Heh, I could say the same thing. You came in later and replied to several of my comments. ;-)

> anxious when performing under pressure

I have to repeat that I'm talking about job interviews specifically, which is not the same as "performing under pressure". Job interviews are unique in many ways and very different from having a job. This ought to be psychology 101 but somehow many techies don't understand it. The pressure of an on-the-job emergency is not the same as interview pressure. They are psychologically distinct.

Anyway, I wouldn't make the general claim that experience makes somebody anxious in job interviews, but I would make the general claim that interview tests often don't actually test for the thing the designers think they do. How exactly does one test for the value of experience, for example?

One thing worth noting though: people who have been out of school and on the job for many years may have become unaccustomed to taking tests, especially if they don't interview for jobs frequently. Fresh college grads are fresh from taking tests constantly in school. So test anxiety can start to affect long experienced people just because it's no longer natural for them compared to when they were younger and in school or new on the job market.

Fair enough.

There definitely can be and often is bias everywhere since people tend to gravitate towards others who are similar to them (i.e. young people might be biased to hire young, old people old, etc.) I can't imagine any reasonable team to take on somebody based on their CV alone though, you're almost guaranteed to hire somebody who doesn't match their own description. I've seen a couple first hand myself.

As I replied to you on the other comment, if you personally have enough options, it's actually better to let some crap companies to have subpar interview processes so that you can detect and avoid them before you take the offer. Pointing out subconscious/unintentional bias is useful though, everyone is better off.

> I can't imagine any reasonable team to take on somebody based on their CV alone though

If by CV alone you mean the mere piece of paper (or pdf document), then I agree. I definitely think candidates should be vetted. I just think that the typical way tech companies vet candidates is awful. I agree with the OP who said "don't have 500 interviews". There's a huge difference between accepting a CV without question and what tech companies effectively do now, which is completely ignore the CV and rely entirely on coding tests for evaluating candidates.

Mass interviews are unusual for many jobs outside of tech. Maybe you just interview the few top candidates. This seems like a better system, and it wastes less time for everyone, both the employer and job candidates, most of whom are necessarily rejected, if there's only 1 opening. Fewer interviews, more time vetting individual candidates. You don't have to do multiple rounds of interviews, and you don't have to do "screening". Start with your few top candidates and work your way down.

If you insist on having a coding test, do it as the last step in the hiring process, not as the first, "screening" step. Job candidates would be much less anxious about coding tests if they knew it was the last step before hiring rather than the hurdle they have to overcome to even get to talk to someone conversationally, and if they knew they weren't competing in the coding tests with a hundred other candidates. Verify, don't screen.

It's also the case that if you start with your top candidates and work your way down rather than mass interviewing, you have time to research the top candidates, such as looking at their open source if they have some. A typical excuse is "We don't have time to look at the GitHub of 500 candidates." So... don't interview 500 candidates!

I agree with you if the person with 25 years of experience only had it a 1-3 companies, you can worry about stagnation.

Maybe. You could also just be screening out loyal people or people who don't like interviewing constantly. I'll let you hire the people that jump ship every two years.

Reply on news.ycombinator.com