Live data from Hacker News

We Hire the Best, Just Like Everyone Else

blog.codinghorror.com

11–20 of 373 posts

Re: We Hire the Best, Just Like Everyone Else

#11
post #9

tldr: "The most significant shift we’ve made is requiring every final candidate to work with us for three to eight weeks on a contract basis. Candidates do real tasks alongside the people they would actually be working with if they had the job. They can work at night or on weekends, so they don’t have to leave their current jobs; most spend 10 to 20 hours a week working with Automattic, although that’s flexible. (Som…

That sounds like it would scale well for the applicant.

It works well in my experience, but we have only used it at smaller scale. 2-3 days working together with somebody is already way more telling than stupid interviews, buzzword compliant skillsets and unrealistic tests.

Re: We Hire the Best, Just Like Everyone Else

#12
>I think our industry needs to shed this old idea that it's OK, even encouraged to turn away technical candidates for anything less than absolute 100% confidence at every step of the interview process. Because when you do, you are accidentally optimizing for implicit bias.

Now you've got to spend time optimizing your interviewing team. What if one guy just says yes to everything? He might as well not be there but he's still a weight in that decision. What about the opposite where he says No every time? When do you decide he is actually going to say No every time and not worth being part of that decision process?

The more people you put on this decision table, the lower the chances that any one candidate will make it through. But there's an opportunity cost of time/money to that, so you've got another thing to balance.

I think it would be really, really hard for one person to simultaneously impress 10 people beyond a doubt just because of how many people have to reach consensus. And can you imagine the stress of knowing one dude with a nitpicky attitude (that none of the others know about yet) could sink your chances?

Re: We Hire the Best, Just Like Everyone Else

#13
My team does take-home project, a small problem to solve that should take at most a couple hours. It's the same problem for every candidate so we're judging them on common work. It's not real work, it's made up.

In interviewing discussions in places like HN, there is enough pushback against this idea, which is a bit surprising to me. I thought a few hours work was reasonable when I was interviewing. But my point is that just this small bit of work seems already too big for many people, let alone working 10-20 hours a week moonlighting. It's cool that they pay for it, but that pay is pretty insignificant in the overall scheme of things. (Also, now you're not really interviewing, you're doing client work. Different relationship.)

When I have tried moonlighting in the past I found the stress incommensurate with the additional income. So I don't think I would do it as an interview, personally.

It sucks that companies take a risk during hiring. But I'm not really interested in making that my problem, as a job seeker.

Re: We Hire the Best, Just Like Everyone Else

#14
post #4

I really agree with the sentiment of audition projects. The interviews I have enjoyed the most all involved some sort of take-home project, although I have never been paid for one. On that note, how expensive does it become for a startup to dole out many interview projects across a wide range of candidates? I think a better strategy would be to conduct preliminary interviews and then based on some granularity, assign…

Audition project just substitute one bias for another. They also presuppose that those who will actually work with the candidate are somehow more enlightened and above bias than the standard interviewers.

Re: We Hire the Best, Just Like Everyone Else

#15
post #4

I really agree with the sentiment of audition projects. The interviews I have enjoyed the most all involved some sort of take-home project, although I have never been paid for one. On that note, how expensive does it become for a startup to dole out many interview projects across a wide range of candidates? I think a better strategy would be to conduct preliminary interviews and then based on some granularity, assign…

Here's a problem with audition projects: I would never be able to take part in one, since my current job contract has a clause forbidding me from simultaneously working for any competitor in the same field. (This is a German job contract; I don't know if such clauses are common in the US or elsewhere, too.) So in order to work on an audition project, I'd have to be unemployed already.

Unless it's not a competitor in the same filed.

Re: We Hire the Best, Just Like Everyone Else

#16
post #6

This might be an unpopular opinion on HN, but another example of this problem manifests in the sheer volume of companies that require a HackerRank screening test these days. Apart from the clumsy web editor, recording someone under a time constraint means nothing in the larger context of how a programmer's actual work happens. Sigh. By using such automated tests, companies want to identify top talent by taking a shor…

Couldn't agree more with this. Also Codility tests in which fresh graduates do very well and more senior engineers fail, because most of the academical/theoretical/generic stuff is forgotten since it's rarely needed in the real world.

Re: We Hire the Best, Just Like Everyone Else

#17
There is no "the best". There are only people. Some people work wonderfully at one shop, but horribly at another. This can have nothing to do with their skill or even their personality. Each working environment is unique as is each candidate. You simply cannot tell which will work perfectly until long after someone is properly hired.

So why bother interviewing? The goal is not "we only hire the best". Hiring is only half the battle and a short term goal at most. The long term goal should be "we only retain the best". Sometimes that means firing people, but more often it means going to the mat to keep the people who work best: Treating people like human beings and, most importantly, paying them. Pay well and the good people will not leave. Actually fire those who don't work out in the long term and even they will do nearly anything to stay.

Or do what most startups do. Pay next to nothing. Treat everyone like widgets in a great machine. Fire only those whose admit having a life outside of work. And hire only those who share similar opinions on ultimate frisbee because culture!

Re: We Hire the Best, Just Like Everyone Else

#18
post #4

I really agree with the sentiment of audition projects. The interviews I have enjoyed the most all involved some sort of take-home project, although I have never been paid for one. On that note, how expensive does it become for a startup to dole out many interview projects across a wide range of candidates? I think a better strategy would be to conduct preliminary interviews and then based on some granularity, assign…

Audition project just substitute one bias for another. They also presuppose that those who will actually work with the candidate are somehow more enlightened and above bias than the standard interviewers.

That seems more like a question on how the candidate's work is measured.

While there's always some bias, I think it shifts quite a bit once you move from an artificial "interview environment" to a work environment.

Plus as has been said, since they candidates are working with the people they would be working if hired and that relationship doesn't work out, then maybe it's better to keep looking for someone else.

Re: We Hire the Best, Just Like Everyone Else

#19
post #8
post #4

I really agree with the sentiment of audition projects. The interviews I have enjoyed the most all involved some sort of take-home project, although I have never been paid for one. On that note, how expensive does it become for a startup to dole out many interview projects across a wide range of candidates? I think a better strategy would be to conduct preliminary interviews and then based on some granularity, assign…

While it is indeed possible to cheat on the project itself to have any value that project gets followed up with an interview in which its discussed, and I think most people would trip up quite quickly if asked to explain in detail the process of developing some software they didn't in fact write.

Exactly. Asking the candidate to outline the solution, the issues encountered, and the rationale behind the design should quickly illuminate any funny business. Depending on the scope of the project, you could even have them push code to a company repo real-time as they make progress. Obviously VCS histories can be rewritten in some circumstances, but I think it would help give a little more granularity as to how the work was done.

Re: We Hire the Best, Just Like Everyone Else

#20
tl;dr: The only measure of on the job performance is on the job performance. The only thing that matters to your company is also on the job performance. Why not focus on that?

It is utter voodoo.

I came up in the late 80s. Interviews were maybe 2 hours long. You might be handed a piece of paper with a problem to aggregate some information distributed in a few different arrays and print them out. You know, write some for loops and a few logic statements.

Then you talked about the job. This is the job. Do you want to do it? What do you have to bring to us? Do you have a lot of experience and want to lead, or not a lot and want to learn - we'll adjust position and salary.

And you put a team together. Some were great, some were okay, some needed to be let go. In total, all sw people had a job.

Contrast that with today's voodoo, where proxies are weighed more than on the job performance (the only thing that matters).

In the end, you put a team together. Some are great, some are okay, some need to be let go. In total, all sw people have a job.

It's all exactly the same, except the absurdity of interviewing by proxies. It's simple logic - he average of everything is average. Why have we abandoned logic?

So many things matter more than remembering red-black trees from your midterm (I'm 49. I TA'ed a graduate level algorithms class back in the day, but I don't happen to remember it. That's what books are for). Like being able to run a project. Being able to write documentation. Being able to enter a room with a combative and upset client and keep the business. Being able to mentor your colleagues. Taking ownership. Leading by example. Ability to learn. So many things that are not even discussed in the current interview environment.

I've watched companies spin their wheels for months, rejecting perfectly good people, looking for that mythical person with exactly the right, esoteric combination of skills, who, for some unknown reason, wants to stall their career and get hired into a position where they learn nothing because they are expected to know everything already. People that are eager to learn? No, sorry, not rock-star material. All these prior successes mean nothing, no one could possibly learn a technology or new algorithm, right? And, more than once I've had people get downright snide about it. I'm sorry that you misread my resume, contacted me, and I didn't have that absurdly specific combination of skills. My fault, right?

Y'all have lost your minds. :) Which is okay, we're an eccentric bunch, but jeez, let's inject some reason and introspection into it. The faces are different, but the talent is no different than the 70s and 80s. There's zero evidence that any of these interview techniques are reliable. There's tons of evidence that interview techniques are horribly biased in many ways. Just stop.

Post reply on HN