News for hackers. Stuff that matters.
Re:Ask HN: What have you found the most useful interview que... by EnderMB on Monday February 04, 2019 @12:28PM (#19078363)
In interviews I like to try and get a reaction, and the best way to do that is to sit down with a coffee or two (or three) and just chat about their experience over the last few years, in a judgement-free zone - dev to dev.

In terms of an actual question, the only useful ones are ones that come from this chat, because it's where we have a back-and-forth conversation about tech decisions, what issues they had, what they had to overcome, and what decisions that particular dev made. My goal is to get a reaction, negative or positive, and dig into why that situation happened, what they did to overcome it, and what they'd do with hindsight.

Not only does it make an interview more interesting for all involved, it tells you more about the candidate than a quiz would. You learn what they do under pressure, what pisses them off, and what experience they have for the kind of work you do, because you've been using your experience against theirs in conversation. Sometimes, to make it fair, I tell them (without dropping client names) issues I've had recently, and I see if any of the solutions were raised by the team, or are ultimately the solution we went for.

I've had enough success with this approach that (when I did interviews) this was my approach:

1. A really basic take home test, probably in a language they don't use, just to see if they can write any code. Literally something that takes an hour to do. A favourite of mine was to get them to write an application in a given language to hit an endpoint, with a given code and their email address. When done correctly, an email is sent to our HR team, who then set up an interview. Based on our logs, you'd be surprised at how many candidates couldn't do this - by which I mean had clearly worked on the test, submitted something, and had failed to send a PUT request with the correct data.

2. A relaxed interview, like above, in neutral ground like a coffee shop. I use the walk from office to coffee shop to handle company questios, and the standard interview stuff,

Re:Ask HN: What have you found the most useful interview que... by kstenerud on Monday February 04, 2019 @03:41PM (#19080412)
I would have failed your test. I'd have assumed that you didn't want your server inundated by requests, set up my own server to verify what my client was doing, and then submitted the code.

Even if you specifically told me that I must actually hit your server, I'd still test on my own server, and then possibly forget to change the endpoint before submitting (no way to unit test that). Shit happens, and you lose someone on a technicality.

Also, what if your server stops working? Now you're wasting people's time.

Don't rely on magic; it eventually bites you in the ass.

Re:Ask HN: What have you found the most useful interview que... by doteka on Monday February 04, 2019 @06:49PM (#19081830)
Not OP but sounds like the test is working then.

> I'd have assumed...

I would expect a good candidate/colleague to ask instead of assuming things.

> And then forget to... > Shit happens...

If someone can't manage to complete a simple assignment without forgetting half the steps, I don't want them pushing code to production.

And a service with one http endpoint having reasonable uptime seems very far from magic.

Re:Ask HN: What have you found the most useful interview que... by EnderMB on Tuesday February 05, 2019 @04:22AM (#19083923)
Yeah, you probably would have failed, since you need to hit the endpoint to get the HR email address as a response.

Literally, the only point of the exercise is to get that email address. It says that much in the description of the task, alongside raising a GitHub issue if the server is down. It was hosted on Azure, and had our uptime tool checking it anyway. If it was down, there'd be a badge on the repo homepage saying it was down. It happened once, and I invited the person to the interview regardless regardless.

I say this much when we send over the technical test. It's like the weigh-in for a boxing match. If they want to showboat, it's on them, but all I care about is whether they can follow a simple instruction with minimal code in the time we said it'd take. I couldn't give a shit what fancy tricks you do in your code, because it's code you've written without time constraints in your own time. The whole reason for chatting about past experience is to see what you do in a REAL development situation.

Re:Ask HN: What have you found the most useful interview que... by insomniacity on Monday February 04, 2019 @04:37PM (#19080882)
Not GP, but I would say that lack of attention to detail is a negative mark. Depending on GP's industry or team, that could be a dealbreaker.

If you can't connect to the server, or it returns a fatal HTTP error, a reasonable candidate would reach out via other means (assuming they aren't deliberately hidden or unavailable).

Re:Ask HN: What have you found the most useful interview que... by EnderMB on Tuesday February 05, 2019 @04:29AM (#19083946)
We had covered all of these eventualities. We had an uptime badge on the repo to say whether it worked or not, and we encouraged developers to tag us on a GitHub issue or email us directly if something seemed wrong.

We were also keen to stress that this should take far less than an hour of their time. We didn't want to see dependency injection, metaprogramming, or anything overboard that showcased a developers knowledge - this is purely just to see if they can deliver code that matches a requirement in the time provided.

In the case where we had downtime (once in 18 months), I just forwarded their email to HR and we spent some extra time going over code in the sit-down interview.

Re:Ask HN: What have you found the most useful interview que... by ManlyBread on Tuesday February 05, 2019 @10:15AM (#19085973)
If I would see that the server is unreachable I wouldn't contact anyone, I would interpret that as a red flag. The company would have no problem with deeming me unfit to work for them if I fail to complete the task - why should I consider working for that company when it's their fault that I'm unable to complete the task? What if this situation illustrates how the company works on the inside (no one takes care of the environments, things are constantly broken)? Is it worth it to find out?
Re:Ask HN: What have you found the most useful interview que... by Maert on Tuesday February 05, 2019 @11:17AM (#19086669)
That's a lot of jumping to conclusions. A simple "Hey guys, the server seems to be down, not sure if you're aware. Can you let me know when it's back up so I can submit my request, as intended?"

Shows understanding, shows willingness to reach out and solve problems, shows ability to delegate responsibility.

Re:Ask HN: What have you found the most useful interview que... by ManlyBread on Tuesday February 05, 2019 @05:49PM (#19090834)
My approach is merely a reverse take on "it is better to pass up a good candidate than to hire a bad one" that most companies seem to utilize. I'd rather refrain from continuing the interview (even if the company wouldn't end up being that bad) than wasting several months in a bad place.
Re:Ask HN: What have you found the most useful interview que... by world32 on Wednesday February 06, 2019 @08:01AM (#19094746)
You sound like exactly the kind of person this test is supposed to weed out: making unfounded assumptions, over-complicating things with a resentful attitude and not even completing the very basic thing that was asked of you.
Re:Ask HN: What have you found the most useful interview que... by bjourne on Wednesday February 06, 2019 @04:59AM (#19093939)
> 2. A relaxed interview, like above, in neutral ground like a coffee shop. I use the walk from office to coffee shop to handle company questios, and the standard interview stuff,

Perhaps that is relaxing to you and most people but not everyone. Coffee shops can be very distracting because you hear other people talking and walking by and some even play music. Perhaps you want to screen out such candidates anyway so it doesn't matter much. But you are also screening out a lot of skilled developers who don't function well in distracting environments.

Re:Ask HN: What have you found the most useful interview que... by EnderMB on Wednesday February 06, 2019 @11:56AM (#19096891)
I agree completely, and this is actually the main reason why I moved it to a coffee shop, because my office was lucky enough to be directly underneath a quiet coffee shop with comfortable seats, lots of table room, and most importantly a relaxed and quiet atmosphere.
Re:Ask HN: What have you found the most useful interview que... by world32 on Wednesday February 06, 2019 @07:45AM (#19094656)
> and some even play music

Sorry to be harsh but if you are unable to have a conversation in a coffee shop its unlikely you will make a good developer, regardless of how well you can code.

You don't have to be a loud-mouthed extrovert or "heart-and-soul-of-the-party" type but social skills are necessary for most dev jobs.

Re:Ask HN: What have you found the most useful interview que... by bjourne on Wednesday February 06, 2019 @09:32AM (#19095302)
I've been working as a developer for close to 20 years. :) So, kindly, go fuck yourself.
Re:Ask HN: What have you found the most useful interview que... by world32 on Wednesday February 06, 2019 @10:07AM (#19095605)
Ok
Re:Ask HN: What have you found the most useful interview que... by _mrmnmly on Tuesday February 05, 2019 @04:19AM (#19083911)
Same here. I've recently been put on the 'other side of the table', first time as a recruiter, and tried to be someone like you've mentioned in your comment. If someone would like to learn more, here's my brief blog post about the whole process: https://lukaszkups.net/notes/on-the-other-side/
Re:Ask HN: What have you found the most useful interview que... by world32 on Wednesday February 06, 2019 @08:24AM (#19094855)
Its interesting to see how many people seem genuinely annoyed at this interview technique. I think that shows the value of it - this test does not seem aimed at testing how much of a "leet" coder you are but whether or not you can get a job done.