Tech sector job interviews assess anxiety, not software skills: study • Hacker News Board

Re: Tech sector job interviews assess anxiety, not software skills: study

#981

Post by kmclean » Thu, Jul 16, 2020, 9:14 PM UTC

musingsole wrote: post #947

kmclean wrote: post #944

What are people doing as alternatives to these kinds of interviews? The research is quite clear that technical interviews are worse than useless (literally -- they will select for the wrong traits in your candidate pool), yet they still seem to be the dominant hiring practice. My current company does a few rounds of low-key chatting, some about technical stuff, then a 1 to 2 month trial period with the understanding…

I like that model your company is using. Though there is a major drawback: your sapping the momentum of your candidates in their job search. If they sign on for the 2 month trial and it doesn't work out, contacts in their job search will have moved on, and they'll effectively have to start over from the beginning. Ultimately, a contract-to-hire strategy overwhelmingly favors the employer's interest.

Interesting, I hadn't thought about it that way. I always prefer a month or two trial run before I commit to joining a company for the long term. I just can't tell enough in a few hours of interviewing whether I'll enjoy working with the people or not. And also you rarely get to meet all your future co-workers in the interview. But then, this has always worked out well for me. I could see it being worse and/or disruptive if it led to a situation of doing one or two month contracts for a really long time.

Re: Tech sector job interviews assess anxiety, not software skills: study

#982

Post by cutemonster » Thu, Jul 16, 2020, 9:33 PM UTC

collyw wrote: post #926

deeg wrote: post #231

Earlier quoted context omitted.

I'm not a googler (but I've interviewed there) and I disagree somewhat with the original premise. I think whiteboarding does filter out bad candidates but also filters out good candidates who fail the anxiety test. Google, with their massive pipeline, probably doesn't care about the false negatives enough to change; they still find enough people to hire. It's the companies with smaller pipelines that need to scoop up…

I used to be reasonably good at algorithmic stuff in university 20 years ago. Since then I hardly ever practice. It's not that I am bad at them, now, but its just so hit and miss whether an answer will come to me in time, you might as well toss a coin. If I was doing algorithmic stuff every day I have no doubt I would be better. Usually what happens is I get out of an interview and a an answer (or a better answer tha…

It seems you cannot see FAANGs perspective on this

There are other people who always fail those coin tosses

Re: Tech sector job interviews assess anxiety, not software skills: study

#983

Post by jwaterhouse » Thu, Jul 16, 2020, 10:23 PM UTC

My 2c. This is purely anecdotal, and based only on hiring about a dozen developers.

The best way to hire good developers is to take them out for coffee/lunch/beer and talk shop for an hour or two. That's it. Make it more of a discussion than an interview. Ask them about the their experience, what have they liked working on, what have they hated, but also tell them about what you've been working on and see what feedback/ideas/questions they have about it. I find fielding their questions and comments just as valuable as asking them my own. Joke about things that have gone spectacularly wrong, trade war stories, that kind of thing.

I think it's a great way to assess whether a dev "knows their stuff" or not. A couple of other notes: - If you're not a dev yourself, get someone who is and you trust to do it. Try and keep it "dev only" if possible, remember this is supposed to assess their dev ability, but is also a good indicator of communication and cultural fit within the dev team. - Try to do this one-on-one or at most 2-1, any more than that and it becomes a panel interview. - If someone is regurgitating theoretical or blog-post-scanned information without much understanding, it becomes more apparent the longer you chat with them. So an hour or two is a good length of time.

We usually also do a quick take home coding test too, but have found the chat/discussion to be better indicator of long term success. As many have already pointed out, I also find the coding tests (whether take home or in person) don't really assess "real world" ability and can often be gamed.

Re: Tech sector job interviews assess anxiety, not software skills: study

#984

Post by katzgrau » Fri, Jul 17, 2020, 12:33 AM UTC

SkyPuncher wrote: post #640

katzgrau wrote: post #559

Earlier quoted context omitted.

In my experience, with C2H you typically earn more in the contract phase than you would with a base salary, which is helpful in covering benefits or out of pocket expenses. Maybe that's not always the case, but salaried employees usually earn 80% or less what their contract counterparts do. And you kind of have to bomb the contract phase to not be given the offer, unless there are other circumstances at play (budgets…

> Maybe that's not always the case, but salaried employees usually earn 80% or less what their contract counterparts do. That's because the employer is covering significant costs on behalf of the employe. A fully-loaded FTE typically costs 30% to 40% more than they're base salary. Contractors need to make more because they're covering all of those things themselves.

I agree, which is part of why C2H isn't as bad of a deal as the parent thread is implying

Re: Tech sector job interviews assess anxiety, not software skills: study

#985

Post by monksy » Fri, Jul 17, 2020, 1:37 AM UTC

jackcosgrove wrote: post #763

Sivart13 wrote: post #521

Earlier quoted context omitted.

"Take home" questions are heavily biased in favor of the employer because they're not required to put anything on the line. An asynchronous test is definitely more flexible for all parties, but it also means the employer can more forward with far more interviews than they would be able to do in person. Leading to more candidates wasting their time doing nonsense work when they have proportionally less of a chance of…

The worst experience I've had with a take home problem was when the interviewer did not even run the program I submitted. As I explained the solution it became clear that the interviewer had not even looked at or run the program. It would have been easy, it was in Python. To top it off the interviewer then sprung a surprise leetcode problem on me that had to work and was required to be written in a language ill-suite…

If you think that's bad. I did my code submission tested the crap out of it. Then found out that the person "couldn't run it". We had a back and forth conversation about it. I even created a docker test bed with the right environment.

mvn clean package worked.

Turns out the guy was running mvn exec:java.

Re: Tech sector job interviews assess anxiety, not software skills: study

#986

Post by lostcolony » Fri, Jul 17, 2020, 3:40 AM UTC

gwbas1c wrote: post #620

lostcolony wrote: post #171

Earlier quoted context omitted.

Why is it a bad thing? It's off the cuff, and likely less filtered than a prepared statement. If they end up somewhere sensible, then your interview process actually worked; you found someone who when confronted with a new problem, is able to think sensibly about it, even in an interview situation.

Because if you're talking without knowing what you're going to say, you're not thinking about the problem at hand. The right thing to do is to narrate what you're thinking about. It's very different than starting a sentence without knowing what you're going to say.

"you're not thinking about the problem at hand."

But OP said he ended up somewhere sensible. He -was- thinking about the problem at hand. As he spoke. He got there in a roundabout fashion, perhaps, and had to think quickly, but it's still -valid-. He's not just bullshitting the interviewer (which is very noticeable and easy to suss out, and never ends anywhere sensible).

Re: Tech sector job interviews assess anxiety, not software skills: study

#987

Post by wallfacer » Fri, Jul 17, 2020, 4:02 AM UTC

twicetwice wrote: post #806

Judgmentality wrote: post #77

Earlier quoted context omitted.

> That and debate gave me the ability to just start a sentence without knowing where it will end and then talk my way to a conclusion that sounds like it makes sense You're effectively advocating for more bullshit. The problem is that this works in the first place - we shouldn't be rewarding it.

You're right, unfortunately that's what it sounds like I'm saying. It's not really what I meant. The idea I was trying to get across is closer to "being able to communicate what I'm thinking about as I'm thinking about it, and verbalizing that as a coherent sentence rather than just fragments." Ironically, I wish I had communicated that better :)

I’m a person who finds out what I’m thinking as I speak. Iterating towards an idea with confidence in each turn. Even if you end up in the vicinity of where you started, problem not actually solved, you demonstrate how you move through problem space.

Similar to another comment I read comparing interviews to athletic tryouts and competition, which tries to assess how a body moves through physical space.

Re: Tech sector job interviews assess anxiety, not software skills: study

#988

Post by bananaface » Fri, Jul 17, 2020, 6:55 AM UTC

collyw wrote: post #907

bananaface wrote: post #872

Earlier quoted context omitted.

Take-home tests are stupid because they instantly filter out anybody with options. If I'm being hit up by recruiters, I might come for an interview with you, but I'm not going to waste time with your test. You're going to end up with monkeys. But even beyond that, it's so dumb to re-demonstrate coding chops to every single employer, especially when I have no idea what my chances are to begin with. Are you interviewin…

Yes, I did one recently, got told it should take 2 or 3 hours, so I spent a morning on it. Got failed for a number of the most trivial reasons - it didn't meet PEP 8 (it was two white space errors, as I didn't have pep installed on my home laptop, so I missed them). A load of stuff about not mocking tests (setting up that would have doubled the scope of the project). The spec and said if you don't have time, then des…

Good God. That sounds awful.

Re: Tech sector job interviews assess anxiety, not software skills: study

#989

Post by jjav » Fri, Jul 17, 2020, 7:21 AM UTC

javert wrote: post #309

jjav wrote: post #170

Earlier quoted context omitted.

Genuine programming skill has no correlation to the skill of regurgitating algorithms in a reality-show type setting. Actual development, of real products, isn't done like that. This is the reason why it's so perpetually controversial to put developers through this type of interviews which measure nothing relevant to the job.

> regurgitating algorithms in a reality-show type setting You're not wrong, but this falls under the genus of "companies that suck at interviewing," and thereby hurt themselves (i.e. the company). In other words, it's not an inherent fault of the programming interview. But there is more to it than meets the eye. Some companies actually just are that incompetent. But, to take a particularly successful and egregious of…

> They want the mediocre people who are also obedient. The kind of people who can be > told to study an algorithms book, and will do it, and will regurgitate it, as a > show of obedience, mostly, but also mediocre (but not great) intelligence--- > not great, because if they had great intelligence, they wouldn't waste their > time on that kind of activity.

Perhaps a bit cynical but there is truth to that. A huge company just needs cogs that can do the job and don't think outside the box much.

Re: Tech sector job interviews assess anxiety, not software skills: study

#990

Post by lazysheepherd » Fri, Jul 17, 2020, 8:58 AM UTC

CobrastanJorji wrote: post #115

I conducted a couple hundred interviews for my first FAANG employer, and I was constantly amazed at the percentage of candidates with years of Microsoft or Facebook experience on the resumes who apparently did not know how to program. I always thought, 'huh, guess I know why they quit after 3 years, amazing that they all lasted this long." Then I interviewed for another company and utterly bombed. It became suddenly…

I've done my first programmer hiring for my startup several years back. Thankfully I've came across Guerilla Guide to Interviewing[1] article by Joel Spolsky, and that helped greatly.

I've interviewed 6 candidates within a week, and hired one. Never regretted of the outcome.

I've set up a test code so: an input variable, comparison variable, and an empty function. Function takes input variable as an input, and output of which is compared with comparison variable.

Interviewee would be asked to fill in the body of the function, and play with it until function gives the expected output. I've set up a computer with two screens, put IDE on one, and Google on the other. Two chairs, and some coffee.

First I've chatted with the interviewee for about 10 minutes, tryin to make them comfy as possible. Then before the test, I've stressed very much that I am also a programmer, and I know how awkward it is to be coding while someone watching, and that'd alone would make me do silly mistakes, so I was expecting same from them and that was completely OK.

Afterwards, I've encouraged them to use Google in plenty, and feel free to stay silent or explain as they go.

So I got to assess their fluid intelligence, their ability to break the task down and progress efficiently, their English proficiency[2] (if they use Google in English), their usage of keywords, their choice among in stack overflow responses, etc.

At the end I've rejected a guy with 8 years of experience on his CV, and hired a junior. Looking back, that turned out to be an amazing decision, best I could have made, for work went good with the junior and I've had the (mis)fortune of working side by side with the 8yr guy several years later.

PS 1: Task was to write a recursive function to travelse a multidimensional array, and find out whether the first letter of every "value" was a capital letter. PS 2: The work was to maintain and develop mid-scale SaaS project along with me.

1: https://www.joelonsoftware.com/2006/10/25/the-guerrilla-guid... 2: Country's mother language was not English, and English proficiency was not good on average.