Live data from Hacker News

The software industry's greatest sin: hiring

neilwithdata.com

561–570 of 590 posts

Re: The software industry's greatest sin: hiring

#561

Earlier quoted context omitted.

Have you ever seen a resume that doesn't have the candidates address at the top of the page? I haven't. We typically receive hundreds of applications for any one job. We phone interview maybe 20 of them and in person interview maybe 10 of them. Those who we don't hire are never given a reason why we chose someone else. They don't know if it is because they don't have the requisite skills or because they live 2000 mil…

These responses cast serious doubt on some of your earlier comments.

In what way?

Re: The software industry's greatest sin: hiring

#562
post #176

Earlier quoted context omitted.

> It has never ever been the least bit about meritocracy. You might find it as amusing as I did to learn that the word ‘meritocracy’ was intentionally coined to mean more or less the opposite of what you think and how you used it here — meritocracy was a derogatory word intended to highlight the social injustice of thinking that ‘merit’ is somehow fair and can be measured and used to rank people. https://en.wikipedia…

> Wait, aren’t credentials one type of “merit”? Credentials are only a weak signal of merit. They are often gained illicitly and are no substitute for actual ability.

What is your proposed alternative for a hiring process that doesn’t involve credentials? (And have you actually tried it?) How do you demonstrate merit in a practical amount of time? What is a strong signal of merit? What does “merit” actually mean to you?

When I hire people, I care about potential and attitude more than merit. I don’t care as much about what someone has accomplished as I do about whether someone wants to learn, is smart and eager to be part of a team. I’ve seen with my own eyes people that come with a lot of accomplishment and a ton of skill, people who have a lot of merit by any definition, and who are terrible people to hire who cause real damage on teams. I don’t believe that “merit” is a particularly strong signal for job performance.

What are you referring to about credentials being gained illicitly? How often does this happen? Are you talking about degrees? I’ve never met someone with an illicitly gained degree, and it would be stupid to try that; you’d get caught immediately. It’s not as easy as just lying. If you’re talking about degrees, I serious doubt that happens “often”, but feel free to provide some evidence that it’s more than statistical noise or anecdotes. If not degrees, what are you talking about?

Lastly, and maybe most importantly, the meta point here is that “actual ability” is to some degree a socially unjust metric. IQ correlates with family income. Why? Because people with more money get better nutrition, more training, better schools, stronger business networks. The advantages in life are, statistically, a major component of what leads to “ability” in the first place.

Re: The software industry's greatest sin: hiring

#563
>> It's a disguised IQ test

Funny, an IQ test has one of the best correlations with job performance among all hiring methods; one of the few to do better than random chance.

>> Does this person have incredible grit and perseverance? Who cares! Does this person communicate and write well? Who cares! How friendly and amicable is this person? Who cares!

I specifically evaluate all of those things and write them up in the report after a technical interview. I don't know where this person is interviewing, but all of these do matter.

But it doesn't matter how friendly you are if you can't also demonstrate problem solving and coding ability.

Also, what a disaster these threads always are. Depressing.

Re: The software industry's greatest sin: hiring

#564
post #562

Earlier quoted context omitted.

> Wait, aren’t credentials one type of “merit”? Credentials are only a weak signal of merit. They are often gained illicitly and are no substitute for actual ability.

What is your proposed alternative for a hiring process that doesn’t involve credentials? (And have you actually tried it?) How do you demonstrate merit in a practical amount of time? What is a strong signal of merit? What does “merit” actually mean to you? When I hire people, I care about potential and attitude more than merit. I don’t care as much about what someone has accomplished as I do about whether someone wan…

> How do you demonstrate merit in a practical amount of time?

I do think standardized tests do a pretty good job. Test mostly on algorithms/tech related questions and you'll mostly find people with interest in these topics. You're right that people who do well on these types of tests may still be terrible people, but it hasn't been my experience that interviews are good at snuffing out terrible people as opposed to people who aren't like the interviewers (ex. people who grew up in poor environments whereas the interviewer didn't).

> What are you referring to about credentials being gained illicitly? How often does this happen?

Regarding degrees, it's well known for example that the Ivy League have large biases towards legacy admissions and for example Asian Americans are highly penalized due to affirmative action. Any effects caused by these types of biases are immoral to me, maybe "illicit" wasn't the right choice of word.

> IQ correlates with family income. Why? Because people with more money get better nutrition, more training, better schools, stronger business networks. The advantages in life are, statistically, a major component of what leads to “ability” in the first place.

This is not true. In the US at least, the effect of shared environment on IQ is known to be very low by late adolescence. IQ correlates with family income because smarter parents have higher income and pass their genes down to their children. https://randomcriticalanalysis.com/2016/05/09/my-response-to...

Re: The software industry's greatest sin: hiring

#565
post #562

Earlier quoted context omitted.

What is your proposed alternative for a hiring process that doesn’t involve credentials? (And have you actually tried it?) How do you demonstrate merit in a practical amount of time? What is a strong signal of merit? What does “merit” actually mean to you? When I hire people, I care about potential and attitude more than merit. I don’t care as much about what someone has accomplished as I do about whether someone wan…

> How do you demonstrate merit in a practical amount of time? I do think standardized tests do a pretty good job. Test mostly on algorithms/tech related questions and you'll mostly find people with interest in these topics. You're right that people who do well on these types of tests may still be terrible people, but it hasn't been my experience that interviews are good at snuffing out terrible people as opposed to p…

Are you advocating standardized tests as a hiring process? Have you actually hired people this way? How many? How well does it work? Does it actually solve any of the problems the article at the top was talking about, or any of the comments up to this point?

Other than the writing tests, math and science standardized tests fail to predict college and especially graduate school performance very well, and it gets worse for predicting career performance. Standardized test scores absolutely correlate with SES, according to the testing agencies themselves.

https://www.act.org/content/dam/act/unsecured/documents/R160...

https://journals.sagepub.com/doi/full/10.3102/0013189X209021...

If you think Ivy League racial bias is immoral, why do you believe that “merit”, which is racially biased in the US isn’t immoral?

Graduates from Ivy League schools represent only a tiny percentage of people, and what you’re actually referring to is Harvard, not even the whole Ivy League, a lawsuit at one single school brought by people with a political agenda against Affirmative Action. This was your argument for dismissing credentials and claiming they’re a weak signal. Are you reconsidering this point of view? That seems like really thin evidence for the strength of signal that credentials do or do not provide. If you believe that standardized tests are a strong signal, and standardized tests are used for college admissions, then doesn’t it follow that gaining the degree credential is at least as strong a signal as the standardized tests you advocate?

I don’t know anything about the anonymous blog post you’ve linked to, but it’s not a scientific source, nor a meta-study, and it appears to be cherry picking and have an agenda. I certainly wouldn’t blindly adopt the “inferred” conclusions you read there, just because it all seems plausible or convincing to you. Claiming that money doesn’t affect IQ or merit or outcomes doesn’t even pass the smell test, there’s strong evidence that being poor hampers ability, even stronger if we’re talking about extreme poverty.

There are pretty well known, well documented problems with cultural bias, in the US and globally. Financial inheritance and pure financial advantage are real; money can and does overcome the disadvantages of low IQ. Being rich has immense advantages in every way. If you are convinced that social biases don’t affect merit and that being rich doesn’t influence merit dramatically, you’re certain that poor people must be poor due to IQ, and you aren’t at all curious about why some smart people believe “merit” might be a subtle way to perpetuate the ideas of Social Darwinism, then we should probably stop here.

Re: The software industry's greatest sin: hiring

#566
post #522
post #47

> Developer hiring is broken This seems to be a pretty popular opinion. It totally might be right, but it’s not my own experience, so I have a serious question because maybe I don’t know what’s happening out there with most hiring today - what are the broad-stroke outcomes that demonstrate that hiring isn’t working? Are there statistics that show that hiring has problems? All of the reasons given in the article are c…

I would say the issues with interviews are * they're not cheap (although not that expensive all things considered), assuming you spend say 10 person hours on two calls and an on site. * they produce false negatives. on the spot exam style algorithms interviews don't reflect day-to-day work, and they filter out people who might be great at the job, but aren't used to the specific type of stress that interview creates.…

> we can look at most hiring processes and see that they are a problem

I’m not seeing a compelling problem with the outcome. The question is: are companies not able to hire good people? And: are good people not able to find jobs?

Interviews are also expensive for a company. And I don’t really see a clear or compelling argument that interviews somehow cost the candidate money. How much money? If the candidate doesn’t have a job, then what exactly is money or time cost to the candidate? Doesn’t the eventual job with income outweigh all interviewing costs? Isn’t the cost somewhat irrelevant if it’s not egregious, and there’s really no choice if you want a job? Isn’t the alternative, choosing not to interview, potentially way more expensive?

If you’re going to frame it as expensive, I think it’s more than fair to counter with you need to average the cost of the interview over the time period that you’re employed. If you interview for 10 hours, and you work there for 5 years, the cost of the interview is 20 seconds per day.

> they’re painful for the candidate

I agree that interviews have a time cost and that candidates have a finite budget. That means someone looking for a job should invest wisely, figure out how & when to decline an interview, study the companies before interviewing.

This is subjective, and it’s not a solvable “hiring problem”. This is something the candidate has to fix for themselves. Nobody is going to give you a six figure salary and ask you to work with them for many years without vetting you. It will always take time, and it will always be somewhat uncomfortable and emotionally draining. Some people prepare for interviews, and some people like to interview. If you view it as a skill, one that you can learn and improve at, maybe it will become less painful for you. But don’t expect that companies are ever going to do that for you.

Re: The software industry's greatest sin: hiring

#567

My favorite interview was at digg.com. A four hour interview was going great. My dream job. I knew their tech solid. Then the last guy - walked in [ianeure]. He asked: "What is a having statement in sql". I leaped in - rambled on and on how you can filter aggregated sets. His response: "I don't think you know how they work." I sat there confused and concerned. He explained you don't need a group by with a having stat…

As a former digg user, I gotta say you got the last laugh on that one. Checking his LinkedIn it seems he works for what amounts to a modern payday loan (now the users pay with privacy loss and a monthly subscription instead of high interest) website now.

Getting a little too personal there, don't you think?

Re: The software industry's greatest sin: hiring

#568

Earlier quoted context omitted.

That sounds interesting, is there an english source to read about that?

I can’t find anything concrete, each company does it internally. There are some articles where they publish the starting salary based on your university’s “rank”. Try google translate on this: https://www.google.com/amp/s/www.lexpress.fr/emploi/les-atou... It discusses how there are 6 ranks for business schools, and your salary for the first N years will be based on that. Same for engineering. It’s funny that one of…

Google translate worked really well on the article, thanks! That's kind of insane that the school you went to can give you a raise from 30k euro to 40k euro despite having to do the same job.

Re: The software industry's greatest sin: hiring

#569

Earlier quoted context omitted.

You say that as if you were born perfect. As someone of elder years myself (and not just by SV standards), what you are saying is quite obvious to anyone with experience. But when you're young and know everything, it's not so obvious. We too, were brave believers of complex == good in our youth. Well, I sure was. It's also important to learn by failing (aka learn by doing). Like parenting, there are many things in or…

I'm sorry if I made it sound like I was born perfect. That wasn't what I was trying to convey. Yes, we learn by making mistakes, and I've sure made at least my fair share of them. My point is that horking out algorithms on demand is definitely not the same thing as being a competent software engineer, nor is it particularly predictive of the likelihood of developing those skills. Also, I measure my success by the peo…

> I measure success by the people I've helped

As someone of younger years (by SV standards), hearing the authenticity in this thread is refreshing. What path in the world of tech will put one around peers who intentionally focus on real impact on people and business?

Re: The software industry's greatest sin: hiring

#570

Anecdotal evidence: Several times in my life I've had the strange luck of being told the internal workings of a decision after I was rejected for a job. Usually somebody from the interviewers liked me but couldn't convince the others, and subsequently decides to find me on LinkedIn and send me a direct message. The messages have been eerily similar and are usually like this: "Hey Dimi, I want you to know that I think…

Yeah almost anything is possible so one really has to try not to obsess about it too much. I know it's hard. I can't say I'm able to do it either.

Anyway, my anecdote: we were interviewing for a C developer position. The next candidate was in his 50s. Our project leader took a dislike to him the moment he saw his resume. No idea why but I assumed it was either his age or that this project leader knew him and didn't want to work with him for some personal reason.

The project leader arrived to the interview at least 40m late. In that time this candidate impressed me (as the C guy) very much and I said 'hire' immediately. However, as presaged, the project leader said no.

This candidate had no way of knowing what had happened. He definitely could tell I was impressed so he probably assumed it was his age. Just imagine how he must've felt. I probably should've contacted him to let him know how messed up the our process was.

Post reply on HN