To make this more fair, why not let the designer work at the company's office, so the interviewer can actually see how much time it took, and the interviewee doesn't waste more time than necessary?
Less convenient for people who have a different job at the same time though.
You took out the time to go to some office. You just spent an hour talking about your hobbies and attitudes and the company's working culture etc. Then, a programmer asks you to balance a red/black tree. You do.. what? You angrily throw the marker on the ground, cry insult, and run out the building? No, right? So then what?
I see this repeated a lot on HN and I'm not sure I'd have the guts to walk out of an interview. To be honest, I doubt many people do.
How about a phone call or a quick email to ask what is their hiring process, what is the work and how it is usually fulfill?
In the current market you are not a seller, you are the buyer. Why should I consider working for you?
My experience is that hireing descicion is so noisy that there are no reason to invest too much since the rate of sucess is barely increased anyway.
The more picky the employer seems the less reason to apply. It might be a internal candidate allready chosen or they are just hoarding resumés to keep recruiters occipied.
The decision is allready made from the from the first minutes ... but you can ruin it of course.
I don't think memorization of an algorithm is what should make a senior level candidate stand out. Senior people need to be able to lead, mentor, self manage (this is a huge one IMO), and understand the bigger picture. "How do we avoid having to balance this tree at all?"
The role of the senior Engineer isn't to tell you how to balance the tree. You, or anyone you instruct to, can simply ask google that or use a library.
Their role is to ask why we need to balance the tree. You very definitely can't ask google that, and that's why you need to hire a senior Engineer.
"Hi <dev name>! I think we may need to optimize memory consumption of this distinct element counting code a bit; I remember there is some HyperLogLog algorithm which should help here... let's check how it goes... (reading Wikipedia page)[0]"
I'm with your comment's parent: I don't think I'd have the guts to walk out of an interview either, especially for a job I applied to first.
For company who reach me, it is mandatory.
I'm not sure this is the best approach...
Have you no common courtesy? I terminate the process as soon as I find out I'm not a good fit for the company or vice versa, no matter who initiated first. Continuing with an interview when I'm sure I'm going to decline is just wasting everybody's time.
I've heard of companies asking for proof of other offers so saying so and so offered me 2x is a very risky play if untrue.
My point is that it is also risky if it is true, but the other offer is not a job you would take.
I have walked out of interviews as well. Sometimes the interviewer and I discover together that the recruiter we worked with didn't know their job, at all!
More often, though, it sort of was clear from the start of the interview that the company would not be a good fit for me. I would tell them so, and my reasons why I thought so, and leave after thanking them for their time.
If they ask me to do some simple programming task, I ask them if they have looked at my Github repository and what they think they will learn during the simple task they could not from my repository. If they do would have a good answer for that, for example they want to get to know my thought processes while coding, I am happy to oblige. But if they clearly have not looked at my Github repositories, nor have any inclination to customize the interview process, I tell that I do not believe in the validity of the process.
I have also walked away twice during the one-month trial period. Once the company was unable to find a project for me, even though they kept on promising there would be one for me soon. The other time I discovered I was unable to combine working full-time with finishing writing my PhD thesis.
Anyway, I also found that in half of the cases I walked out, the other party did understand and accepted why.
Of course, it is easy to walk away when you know you will find a job soon anyway and even if you would not you will get some support from the government anyway.
If they don't wave you off at the suggestion, you also get to meet everyone and see their work environment.
What I feel is more important is a persons thought process on how they approach a problem and come to a solution.
One hour interview cannot be more than just screening, when split 50-50 to give candidates enough time to ask their questions. It leaves ridiculously small amount of time to the really interesting technical topics. And this discussion is important: CVs are not trustworthy, plenty of people lie or exaggerate the facts — only a live talk can verify the necessary knowledge, experience and skills. It may not be enough, true, but hiring someone after barely speaking to him is unnecessary risk. And hiring someone without necessary practical knowledge and without proving the ability to learn quickly is the same.
Are you sure you meant to say what you wrote? I'm not aware of any place that has figured out how to properly screen candidates in one hour. If you know how to do it, please share, it would help all of us.
Corporations love obsessive-compulsive workers who feel thankful to have the job, are willing to sacrifice personal time towards their business goals and would never dream to unionize or otherwise engage in collective bargaining. They want solitary and competitive workaholics, that's the whole point of 'homeworks'.
I have to wonder if they didn't get the position because they spent more than the advised time for the task.
If that makes them look like assholes, yeah, I can see why someone would feel that way but at least they are clear about their intentions and not insulting anyone's intelligence in the process.
It'd be more useful to know they could execute roughly what I'd expect in 3-5 hours. As in they think quickly, come up with something, get the basics down and are ready to progress or not.
...Of course, if they're doing it at home they could spend 3-5 days on something I'd spend 3-5 hours on and I'd never know...
Maybe the same situation applies in Design somehow? I mean yeah it takes 4 hours to make the product if you happen to have all the graphical elements you will be using already downloaded and organized.
on edit: although I have noticed a disappointment in some places that I have not done a lot of time cleaning up the code after writing it and making everything beautiful and writing a good readme to tell them everything to do and so on and so forth and normally I don't do those things because I spent more than 4 hours because there was stuff that needed fixing and anyway if they wanted all that 4 hours would not be enough.
Coding is like math homework. When your doing the math homework you know when your done, it's very obvious and anyone can check your work and clearly see it's either right or wrong.
Design is like a book report for English class, it's bottomless and can always be improved given more time, albeit with diminishing returns. Design challenges are evil and bullshit, you can poor hundreds of hours into a design challenge and still iterate further.
But people complain that I am not done with the programming assignment if I have not made it DRY enough, or put in tests, or done a readme outlining more advanced than run npm install then do npm run start.
Programming assignments may not be infinitely improvable but they can often be improved, I just don't consider it worth the effort to improve the work before submitting
Agreed, unless they want a good polished solutions, meaning it has been refactored, has decent comments, is well tested, etc.... then it becomes a lot like design where the sky is the limit.
My wife got her first job in the states because of a take home design challenge. Other companies were wary of her because of her foreign degree and no US experience, and she was able to show off in the design challenge they gave her (she spent more time than estimated, but not incredibly much more than that). In this case, it really worked out for her.
If I asked a candidate to spend 4 hours on something and they come back to me with 40 hours of work, it isn't what I asked for. Despite it being potentially amazing, that person ignored the task.
A 4-hour task would probably yield 1-3 mockups. Maybe a launch/splash page, the home leaderboard page, and then a page showing the details of a specific user.
Even these 3 pages would be pushing your 4 hour time limit, but that is really the most I would expect.
So if I was interviewing and comparing candidates and one candidate submits a 40-60 hour project while the other candidates submitted a 4 hour project. I can't really compare the two. I would probably throw out the guy that ignore what I asked them to do by spending 40-60 hours of work on something I only wanted 4 hours of work on.
Also, I do think that some people are able to complete it in a third the time, but I think the more useful evidence is that many candidates complete it and do a perfectly good job, some even with a healthy chunk of time to spare. If it were a totally unrealistic time goal I'd expect we'd quickly notice when no one finished.
I don’t think Google necessarily requested a full end to end product. A nice logo/app icon for the “branding” portion and some UX concepts can definitely be fleshed out in a single work day, especially with zero tech/biz requirements IMO.
Maybe you’d need to take an extra day just to brainstorm first... but actually pushing pixels, if this designer compressed his output a bit, could have happened in 6 hours.
I had an interview once where we were on the phone, I got the task, and they told me they'd call me back 2 hours later.
Might be easier if the office is far away.
My last company they flew me from another continent to do a 3-days assessment, I worked with the team on a legitimate feature(that would be trashed after as they couldn't use the code because I was unpaid). I liked it, I met the team, I interacted with them, I did the task and got a job offer in the end.
All engineers at the company highly valued the 3 days assessment because we got to select great people to work with, but we were getting too many candidates giving up or false negatives, they eventually cut it down to 1-day despite our protests.
Honestly it's stuff like this, that has me contemplating giving up this career path because there's just no more room for people in their 50s with 25 years of experience in information technology, who aren't willing to sleep at the office, drink your "insert trendy microbrew here" on tap, and suffer through a ludicrous ask like "we want you to work for three days for absolutely nothing to see if you're a good fit".
If you are traveling internationally, then you have to add 1 day of travel on each side of your trip. Plus the 3-day "interview" or challenge or whatever. So that is 5 straight days of interviewing for a job. That is actually asinine.
If you have a current job (most software developers do, especially the good ones) then that is an entire work week that i have to request off at my current job to go to unpaid work with no guarantee of getting a job afterwards.
I can't believe someone has actually done this. It is crazy!
The author of this article fucked up when he tried to pass off 5 days of work as an example of what he could accomplish in 5 hours. I am glad he was able to find inspiration from it, but for the success of his startup I hope he also is able to understand other, better ways he could have approached the interview task.