Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

You can’t fix diversity in tech without fixing the technical interview

blog.interviewing.io

Post reply on HN
Print view View original

141–150 of 504 posts

Re: You can’t fix diversity in tech without fixing the technical interview

#141

Post by Manishearth »

user5994461 wrote: post #128
Manishearth wrote: post #91

Earlier quoted context omitted.

This is their point exactly. "Being good at technical interviews" is not the same thing as "being a good programmer". You can be bad at interviews but good at programming. You can be good at interviews but bad at programming. Turns out that there's a correlation between the former set and certain demographics, because of unconscious biases in the interview process.

Doesn't have to be the same thing, it just has to be correlated enough.

Of course it doesn't. I talk about this in my other comment (https://news.ycombinator.com/item?id=12859833), many companies are okay with false negatives. And false negatives are okay. Sad, but okay.

However, if your metric for hiring is also correlated with favoring a particular demographic, you shouldn't be using it. False negatives are not okay when you disproportionately target one group.

Re: You can’t fix diversity in tech without fixing the technical interview

#143

Post by T2_t2 »

tuxidomasx wrote: post #129

I think the conversation about diversity in tech often overlooks a key metric, which is the percentage of tech companies that are founded by and/or employ a majority of minorities. For example, if its accurate to say that over 88% of US tech companies are founded by white people and employ mostly white people, then a diversity issue exists in the number of tech companies founded by black people and employing mostly b…

Yeh, I'm not sure that's a great idea. Companies fail a LOT - startups especially. We look at the successes and think "this is the way to do it", but I wonder if the failures would be felt most starkly by the poor. If Zuckerberg failed at FaceBook, he was still Ivy league educated from a middle class family. That safety net really helps take that risk.

I'm not saying stop people doing it, but I'm not sure encouraging more high risk taking amongst the most vulnerable is wise.

But I agree with your central idea - diversity of options, beacuse diversity, to me, is not forcing every environment to be as diverse as possible, it is about a diversity in the cultural landscape. As a reductio ad absurdum analogy, diversity isn't forcing mosques to server beer and bacon and forcing pubs to have alcohol and pig free sections, it is having a world where there are both Mosques and pubs, and we can all choose what we prefer.

Re: You can’t fix diversity in tech without fixing the technical interview

#144

Post by balls187 »

pj_mukh wrote: post #94
balls187 wrote: post #36

Indians of Asian descent are a minority in the US, but well represented in Software Engineering. Further, there are plenty more Indian, Chinese, and Russian female engineers, than there are American (of european, aka White) female engineers. I am less inclined to believe there is systemic bias against women, and minorities, and instead a culture difference between groups who are well represented in engineering, and g…

That's because Indian (particularly) men consider engineering to be a safe choice (source: am Indian). Anecdotally, speaking to women (esp American), they find engineering to be categorically unsafe from a career progress perspective. You may not 'see' the bias, but they definitely feel it. The effect is measurable. I am no expert, maybe it has something to do with maternity leave regimes or the interviews (like OP m…

My wife is white, PhD in engineering. She choose the field because that is what she was excited about and wanted to study. Still works in Tech. Is a mother. Not so great maternity benefits though.

Re: You can’t fix diversity in tech without fixing the technical interview

#145

Post by foolfoolz »

trhway wrote: post #30
divbit wrote: post #13

Earlier quoted context omitted.

That is a pretty huge statistic, - what also pops out to me is: "stop practicing ". This seems to imply that these technical interviews are skills that you don't really gain on the job. I know that personally would 9 times out of 10 spend time on a cool / beneficial project rather than making time to practice trivia problems.

>This seems to imply that these technical interviews are skills that you don't really gain on the job. the technical interviews skills are gained, at least in my experience, as result of the technical interviews - which normally implies, and did happen in my case, high rate of interview failures initially. Taking failures hard would naturally lead to Catch-22 here.

the biggest catch 22 for me is when you work with someone for years at a job. at this job you have a strong interview process. you reject the same 80%+ of candidates. then you move somewhere else. you refer your friend. they get rejected.

were they a bad employee? did they sneak in the door at the last place? Is there a problem with how we do interviews?

Re: You can’t fix diversity in tech without fixing the technical interview

#146

Post by bit_logic »

This is because the technical interview has completely degenerated into an arms race. It started with good intentions. It used to be just, here's a problem, try to solve it and let's discuss. And at first it was fine. But then knowing Big-O, algorithm, data structure, etc. became a way to "show off" and impress the interviewer. So everyone started to focus on those to impress the interviewer. Then the interviewers noticed and started to state they would only be impressed by more and more obscure algorithms and data structures.

And then the real downward spiral began with the publication of Cracking the Coding Interview. More books were published, a whole industry to support this. Then came the sites like leetcode and HackerRank. Now there's a lot of money supporting and pushing for continuing this stupid process. And the arms race just keeps getting worse and worse. The interviewers expect more and more obscure algorithms and data structures and justify it with "must avoid false positives". This just increases the study time for candidates and they're willing to spend time and money to do it. And the tech interview industry (all the books and sites) are happy to push this arms race since it's more money for them. Now they have enough money to go openly defend the tech interview as the "best practice".

It's not even enough to get the right answer now. You have to do it fast enough or you fail. And now they care about syntax on a whiteboard. None of this used to be true. It all started as a way to see how you approach a problem and discussing it. No expectations that you would correctly get Algorithm ABC and Data Structure XYZ. No expectation that the syntax is correct, pseudo code was ok. No expectation even on a complete working solution, the point was just to see if you can reason about the problem.

All that is gone, replaced by the tech interview arms race. Now it's just a massive speed pattern matching contest to see if you've studied enough to hopefully cover the obscure problem the interviewer will pick and if you can correctly pattern match in time based on the hundreds of questions you did on leetcode. Completely useless and against the original intent of tech interviews.

Re: You can’t fix diversity in tech without fixing the technical interview

#147

Post by bostik »

trhway wrote: post #30
divbit wrote: post #13

Earlier quoted context omitted.

That is a pretty huge statistic, - what also pops out to me is: "stop practicing ". This seems to imply that these technical interviews are skills that you don't really gain on the job. I know that personally would 9 times out of 10 spend time on a cool / beneficial project rather than making time to practice trivia problems.

>This seems to imply that these technical interviews are skills that you don't really gain on the job. the technical interviews skills are gained, at least in my experience, as result of the technical interviews - which normally implies, and did happen in my case, high rate of interview failures initially. Taking failures hard would naturally lead to Catch-22 here.

> technical interviews skills are gained

Oh yes, and that applies to both sides of the table. Interviewing skills are learned as well - and they can be taught. (A friend is proud of the fact that he's had actual interviewing training.)

Interviewing is not that hard, but it sure as hell is a difficult thing to master. I'm reasonably good, but still nowhere near as good as I'd want to be. Nonetheless, it's still a skill that can be learned and improved.

The irony is that it'll be easier to improve one's interviewer skills than those of being interviewed. The reason is that if you're conducting technical interviews, you'll be doing it as part of your job, and any reasonably good interviewer will have his or her calendar full of learning opportunities.

Candidates will have less opportunities and due to the nature of the experience, are likely to be demoralising to boot.

I recently wrote a post about things I've picked up and learned when doing technical interviews: https://smarketshq.com/notes-on-interviewing-engineers-a4fa4... ; when judging the experience from HN posts, I cannot help the feeling that most engineers are subjected to pretty horrible, even recurring interviewing experiences.

Re: You can’t fix diversity in tech without fixing the technical interview

#148

Post by throwaway420 »

I'm waiting with great anticipation for the politically correct outcry to solve the problem of not enough women being represented in the mining, construction, sanitation, logging, or deep sea fishing industries, or for there be a popular movement to get more men into teaching or nursing where they're also dramatically underrepresented there.

Let's see if there's any intellectual honesty here, or maybe there's something else behind this politically correct movement?

Gee, I wonder why certain people are trying to pu$h this so hard? You can ea$ily gue$$ what thi$ expan$ion of the labor $upply i$ really about if you think long and hard about it.

Re: You can’t fix diversity in tech without fixing the technical interview

#149

Post by xorgar831 »

I've seen a number of tech interviews where the interviewer has played down candidates that were better than them (consciously or not). I've also been a interviewee many times and sensed a one-upmanship from the interviewer(s), and subsequently avoid those companies. I could easily see this also being another factor if someone isn't in tune with the norms of the industry to call BS and walk.

Re: You can’t fix diversity in tech without fixing the technical interview

#150

Post by avasylev »

DigitalSea wrote: post #130

I consider myself a competent developer, but I have no academic background as I am self-taught. While I have had a steady supply of work, because of this, I will most likely never get to work for a big company like Microsoft, Google or Facebook because I wouldn't pass their technical interview. I lack the understanding of core computer science principles and I am okay with admitting that, I don't think it makes me a…

I'm one of the people who ask those algorithm/white board questions. I've read a lot criticism on this topic, here're my reasons to stil keep asking these questions:

1) Working on algorithm problem requires very little in common between interviewer and interviewee. I'm C++/C#, I won't be able to give Ruby dev a real world task and evaluate it properly. Algo/data structure problems are universal and looks similar in any language. In my practice I completely failed to make any conclusions by talking about previous experience. It was not clear to me what person actually did vs what was already on the project, may as well just fixed couple UI bugs on the complex project. This part might be might personal failure as interviewer, but it was it is, and I've seen enough developers with years of experience who barely code. So unfortunately having great resume doesn't prove anything.

2) Everyone knows that big tech companies ask these questions. Knowing that you didn't take to prepare, says one of two things to me: you don't find it interesting topic/usefull or you don't have enough grit to go thru that. When I prepared for the interview I went thru two algo books, it took some time but it was easy for me, I did enjoy the learning part, yes it's not something I use every day. But learning things is a trait of good developer, knowing fundamentals if part of that. Some folks found it useless, but since it's known fact that's being asked on interiewes, motivated developer would spend time and learn it. Over the years at work, we have to learn things that are not interesting and that I'd rather choose not to know at all, same as algorithms to some people. If you can't make yourself learn algorithms to pass interview, likely you wont learn things for the work too.

Post reply on HN