Are Whiteboard Coding Interviews Just Testing For Social Anxiety? (theregister.com) 196
An anonymous reader quotes The Register:
People applying for software engineering positions at companies are often asked to solve problems on a whiteboard, under the watchful eye of an interviewer, as a way to assess technical problem solving skills. But recent research suggests that whiteboard technical tests — so daunting to job seekers that there are books on how to deal with them — often fail to assess technical skill, according to new research. Instead, they're all about pressure.
In a paper to be presented later this year at the ACM Joint European Software Engineering Conference and Symposium on the Foundations of Software Engineering, researchers from North Carolina State University and Microsoft in the U.S. argue that whiteboard sessions test for stage fright rather than, y'know, coding competency... "A technical interview has an uncanny resemblance to the Trier Social Stress Test, a procedure used for decades by psychologists and is the best known 'gold standard' procedure for the sole purpose of reliably inducing stress." As a consequence, whiteboard interviews may fail to assess coder competency. Rather, the researchers argue, they measure how well job candidates handle anxiety....
In essence, social anxiety took otherwise qualified job candidates out of the running because of the circumstances of the interview.
In a paper to be presented later this year at the ACM Joint European Software Engineering Conference and Symposium on the Foundations of Software Engineering, researchers from North Carolina State University and Microsoft in the U.S. argue that whiteboard sessions test for stage fright rather than, y'know, coding competency... "A technical interview has an uncanny resemblance to the Trier Social Stress Test, a procedure used for decades by psychologists and is the best known 'gold standard' procedure for the sole purpose of reliably inducing stress." As a consequence, whiteboard interviews may fail to assess coder competency. Rather, the researchers argue, they measure how well job candidates handle anxiety....
In essence, social anxiety took otherwise qualified job candidates out of the running because of the circumstances of the interview.
Which is more important â" personality or ski (Score:2, Funny)
Re:Which is more important â" personality or (Score:4, Insightful)
If I had limited time to interview, IÃ(TM)d be quite interested in how a candidate performs under stress.
As a more senior candidate, my response might be " do we have this level and type of job stress often?" then nope the fuck out of there (actually I happen to interview well because I'm lucky in that I get fired up by that kind of adrenaline rather than freeze---I also enjoy public speaking, same ting really). I've had tight deadlines, shit going wrong and stuff on fire and that's stressful but wildly different from interview stress. The only time I've ever experienced that sort of feeling at work is when talking to a very large audience, something few people do.
Intentionally stressing the candidate in an irrelevant way is a dick move that tells you nothing.
Pure coding skills can be assessed in an online test.
In theory yes, but in practice it's a test of how much bullshit a candidate will put up with. I won't do any work outside the interview, and I know plenty of people who won't either. You might have doubts about ow good I am (I won't brag, that would make it seem less likely I'm any good), but I certainly do know very good people with the same policy. We've got full time jobs and lives outside of work!
I've given a hard pass to Google now twice over that, and they approached me both times. That's like WTF, you want me to change what I'm doing for your benefit and now you want me to jump through hoops for you? How about no?
Re: (Score:2)
Intentionally stressing the candidate in an irrelevant way is a dick move that tells you nothing.
It's not irrelevant. It allows you to judge a person's reaction to a sudden and inconvenient situation. Surely you've come across such a situation in your years of work. Contrary to popular belief, work is not one continuous process of milk and cookies. Every so often an event occurs which necessitates rapidly changing direction to complete a task or goal.
Seeing how people react to a stressful situation is
Re: Which is more important â" personality or (Score:2)
The point is that a regular work environment for developers shouldnâ(TM)t have this kind of stress, as a matter of course. It indicates a development organization with serious structural problems. For IT? The situation might be different. IT is a MUCH more fire-driven job than development should be.
For developers, how they work as team members, or mentors, or what their interests are, or how solid their interpersonal skills are, or how good their code-reviewing skills are, or how good their documentati
Re: Which is more important â" personality or (Score:4, Insightful)
If you're expected to work and contribute to a team environment, being asked out of the blue to take on a specific area of work in a specific amount of time can be stressful. So is having to explain your work in front of others. Being able to explain your work in general can be stressful.
If you can't perform under a simple interview where there's only a few people (maybe 3 or so), how are you going to perform if you're in a group of twenty? How are you going to explain something to a group of hire ups who make the decisions based on your recommendation or explanation?
Seeing how people react in unexpected situations is necessary. You could have someone who, on paper, looks fantastic with all the right skillsets and experience, but who can't explain why they would use a loop instead of an array (I'm not a programmer so forgive my simplistic example). If you want someone to have good code-reviewing skills, how about they be able to articulate why a section of code is good or bad or needs some tweaks? Telling someone their code sucks, in a polite but business-like manner, can be stressful.
I do agree that questions about trivia aren't overly useful. They're more bonuses which can relate back to their interests.
Coding is a Minor component of Application Dev (Score:3)
Seriously.
You can get legions of high school kids who can code circles around other people. That doesn't mean they know shit about applications, the businesses they support, or the people using them.
Re: (Score:3)
Re: (Score:3)
Sit me down in front of a computer for the task, let me work in an IDE I'm comfortable with or can learn quickly, have people I can interact with to talk to and bounce ideas off of. And most of all, make it relate to the actual projects I'll be working on.
If you need all that to figure out if a string is a palindrome, I think it's time to look for a different career.
Re: (Score:2)
It's not irrelevant. It allows you to judge a person's reaction to a sudden and inconvenient situation.
No, it allows you to judge a person's response to a certain kind of non sudden, not inconvenient but commonly stressful situation. Interview nerves are a thing, and the nervous candidates are going to be nervous well in advance. And this stuff just doesn't come up often in the software jobs.
Surely you've come across such a situation in your years of work.
Did you not read my post? Yes I have, it's the same
Re: (Score:2)
A musician friend of mine is married to a Sysadmin. Don't laugh. It works.
She equates an interview with an audition, which totally different from playing in an orchestra.
How to make it not stressful and actually skill (Score:2)
It seems to me that if you're applying for a coding job, when you go to a coding interview it shouldn't be too surprising that coding comes up. Which is good, because having the test be unexpected is an important part of what makes a similar exercise stressful.
A company I worked at used a simple coding test towars the end of interviews and I think we could normally tell the difference between someone who is just nervous and someone who is clueless. It's not a computerized test where the machine passes or
Re: (Score:2)
If your company regularly experiences these kinds of "stress events", then your company is broken.
Re: (Score:2)
Nah, just your management. They're failing to manage.
Re: (Score:2)
They're failing to manage.
Whether that's a symptom or a cause, it's still broken.
Re: (Score:3)
Social anxiety stress is not anything like having to deal with sudden and inconvenient situations. There is no similarity at all. The stage fright phenomenon is entirely due to the situation having people staring at you judging your performance at every second.
This is well known among professional performers, some of whom relish the crowd, but many do not -- many orchestra members are glad to be hidden in a pit. Even famous professional performers who normally do well in front of crowds can have stage frig
Social anxiety is important (Score:2)
Inability to work under pressure, and inability to deal with others, is not a characteristic you want.
Re: (Score:2)
Unless there's an overheating nuclear reactor or similar, much of the stress is manager generated. If there is an overheating nuclear reactor, I'm not sure why you are trying to get software developers to fix it.
Re:Which is more important "personality or (Score:2)
Pre-COVID, I'd get pre-screening questions that required some research and an essay that would have taken 4-6 hours to do thoroughly. I turned those down. Short questions like "What's the difference between a hard and soft link?" are fine, but I'm not going to write an essay describing in detail at the network level what happens when I type a URL into a browser.
I've had a python coding pre-screen for DevOps gig which was minimally specified. I had to guess how they wanted the output. I'd never really code
Re: (Score:2)
This had been my experience with coding assignment "interviews". The basis of evaluation is not disclosed and you know nothing about the people who will be reviewing it. I have no problem completing "assignments" as specified, and I even provide a list of assumptions and explanations behind I took the approach I took, but the usual response is "not what we are looking for" - without further explanation. I once was told my solution to an algorithm assignment (in which I achieved all of the additional "nice t
Re: (Score:2)
Whow..generally you look for the opposite. Does the candiate bother to get the requirements clear, or do they dig themselves into a hole? If it's an assignment that you are just suppposed to turn in without any interaction, then you're free to make assumptions and put clarification in comments of course.
Re: (Score:2)
Yup.
It's probably a valuable interview technique. From the sound of it, working at a company that does this stuff involves putting up with a lot of bullshit.
Re: (Score:2)
Re: (Score:2)
Two things needed. To know if they candidate has the skills or not - remember, it is easy to lie on resumes, and easy to get glowing recommendations, even when incompetent. And second, yes, the stress happens on the job. If you can't stand some *friendly* interviewers asking questions, then what happens when the candidate has to talk to customers, executives, or even normal staff meetings when things are going wrong. Communication is a vital part of most jobs, we don't send engineers into an office and
Re: (Score:2)
Whiteboarding? (Score:5, Insightful)
So I've done many interviews, often involving a whiteboard and I've had what amounted to a bunch of very odd discussions online where I felt like we talking across each other about "whiteboarding". Turns out we were.
Apparently some lunatics expect interview candidates to write actual CODE on a whiteboard with a honest to goodness pen and want it to be syntactically correct. This is a colossal WTF. It also explains why I've had candidates try to write code on the whiteboard which seemed really odd to me.
I think it's fine to use a whiteboard to do things you would do in your day job with a white board; discussing high level architectures/diagramming and so on with a peer. Often in actual work, there's an informal whiteboard design phase where team members discuss various aspects and try to hash out a design.
But writing code? What sort of lunatic wants that?
Re: (Score:2)
So I've done many interviews, often involving a whiteboard and I've had what amounted to a bunch of very odd discussions online where I felt like we talking across each other about "whiteboarding". Turns out we were.
Apparently some lunatics expect interview candidates to write actual CODE on a whiteboard with a honest to goodness pen and want it to be syntactically correct. This is a colossal WTF. It also explains why I've had candidates try to write code on the whiteboard which seemed really odd to me.
I think it's fine to use a whiteboard to do things you would do in your day job with a white board; discussing high level architectures/diagramming and so on with a peer. Often in actual work, there's an informal whiteboard design phase where team members discuss various aspects and try to hash out a design.
But writing code? What sort of lunatic wants that?
People do that? That is insane. Informal pseudo-code that describes the idea, yes, but being able to write actual code correctly without all the tooling you have usually is a completely wasted skill. You should probably discount candidates that can do this because they focused on the wrong things when learning their jobs.
Re: (Score:2)
actual CODE on a whiteboard with a honest to goodness pen and want it to be syntactically correct
Syntactically correct is unreasonable, and I've never had anyone ask for that. Whiteboard coding is partially a coding test, but more of a communications test. I ask candidates to code simple problems like "reverse a string" or FizzBuzz or just to write a SQL inner join. These questions are so trivial they don't measure someone's coding ability. Interviewers ask those questions as a screener to see if this person a complete fake who just googled a bunch of terms and showed-up. The article asks them to
Re: (Score:3)
If the applicant's CV hasn't been adequately reviewed/screened before the interview stage, someone isn't doing their job.
Agreed. But that has nothing to do with a whiteboard or a coding test. If you have interviewed hundreds of people then you know that reviewing the CV doesn't tell you if they can code. If they are recent grads they may not have a CV. If they do, there are lots of programmers with 10 years of copy/paste coding experience who can't solve a novel problem. I work for an engineering company and we can't use copy/paste programmers. If it is a web site then the needs are different.
There might be an informal discussion of something based on the applicant's aptitude. They may voluntarily demonstrate something. There's lots of options.
Agreed. Don't think that t
Re: (Score:2)
Re: (Score:2)
We write all our code on white boards. Then we make photos and fax them to India, where an IT support person reads them back to us over the phone. We then use text to speech software to translate it into a compilable file. Works great, everyone should do it!
Re: (Score:2)
...and that should have been 'speech to text'. Damn, some way to kill my own joke :-(
If social anxiety is that overwhelming..... (Score:4, Interesting)
If social anxiety is that overwhelming to you, then you cannot present your work or defend it when challenged. Your career options are severely limited, and you're quite likely to have your good work wasted because you cannot present or champion it nor can you effectively climb the corporate ladder to leadership positions.
Fortunately, social anxiety is treatable. See https://www.ncbi.nlm.nih.gov/b... [nih.gov] . The last time I had a candidate blowing the whiteboard part of the interview, obviously flustered, I did let them know we'd had issues with their presentation, and even recommended some training.
Re:If social anxiety is that overwhelming..... (Score:4, Insightful)
Re: (Score:2)
Really... you can train away autism? Not saying that was the case in the example you cited, but it can often be outwardly indistinguishable from social anxiety.
Also, ever hear of these things called online coding reviews? They aren't social, but they certainly still give a person the opportunity to respond to criticism.
Re: (Score:2)
Re:If social anxiety is that overwhelming..... (Score:5, Interesting)
I have never had to present or defend my work in a way that was as stressful as a whiteboard interview. Particularly because by the time I was prepared to present and defend my work, I'd actually had the time to do it, understand it, and be able to answer questions about it.
I once did a whiteboard interview and the interviewer asked me how to code up a Fibonacci sequence solution. I'd just been reading up on that a couple days before, so I jumped straight to the more efficient solution of storing the last two values and adding them up, rather than dithering about with the naive, recursive solution. (My solution wasn't that fancy either, but I'll note that before I read about it, it had never occurred to me because literally every time anyone talks about Fibonacci, the recursive solution is the one they show and then they stop.)
Well, that wasn't what the interviewer expected and he basically questioned my choice of solutions. Not because it was wrong or bad, but because he'd expected to be able to start with the naive solution and dither about asking me for better ones. Rather than accept that sometimes you can just jump straight to a good solution, we had to go back and cover the garbage ones too. It was nonsense.
I didn't get the job, and in retrospect, I'm glad I didn't. Not just because of the interview and what it told me about how work would be there, but that's certainly part of it. It was a good lesson to learn.
Re: (Score:2)
I've been in an interview that was going pretty well but ultimately went over the allotted time. I agreed to go over and they asked me to write a script. I drew a complete blank on how to do it and said so. I'd have to do some research to figure out the best way to do it. That ended the interview.
Afterwards, I did a simple google search and there's UNIX command that does most of the work. But I didn't know that command, so I couldn't solve the problem in the moment.
I supposed I might have been successful a
Nobody ever asked me to write code on whiteboard. (Score:2)
Re: (Score:2)
Why do you feel that compiling it would be necessary?
And if compiling code were necessary to understanding code, then you would not be able to write any code that you understood in the first place, unless you ordinarily write code that you do not understand and are actually just a copy/paste style coder.
Re: (Score:2)
You did not tell the joke well.
Expecting syntactic correctness (i.e. compilability) is actually a common conceit of these interviews, so the joke falls flat.
Re: (Score:2)
Interpreted languages do not require compilation. Actually _running_ the code isn't necessarily available.
Re: (Score:2)
Interpreted languages do not require compilation. Actually _running_ the code isn't necessarily available.
Basically every language gets compiled these days. The speed advantages are just way too good to not do it. Yes, that includes interpreted languages. The difference is what you compile to.
bruh they asked me to compile on a whiteboard (Score:2)
It was pretty tough, I had to build my own parser symbol table on the whiteboard, translate everything into an abstract syntax tree, and then into assembler (at least they let me use Intel instead of AT&T) and finally into machine code ... for MIPS. Why MIPS?
Fortunately I had memorized the entire MIPS machine language and was able to translate everything down into hexadecimal binary.
It was funny because in the end, that's exactly what I wound up doing on the job! Apparently they had some weird manager w
Re: (Score:2)
It was pretty tough, I had to build my own parser symbol table on the whiteboard, translate everything into an abstract syntax tree, and then into assembler
Pffft, what an amateur. I did all that while riding a unicycle blindfolded and then I saved the file to the hard drive by carefully tapping the platter with a magnetized needle as it spun.
Re: (Score:2)
Explains the rise of soyboy software (Score:5, Interesting)
Re: (Score:2)
Twitter was an inside job. While there were probably ways to have caught him before it got this far, keeping a qualified but silently malicious person out of a public company is near impossible.
Re: (Score:2)
IIRC it was someone who snuck into twitter's slack and got access to the tools that way
Re: Explains the rise of soyboy software (Score:2)
Glib, uninformed, and factually wrong edgelord bullshit. Coding practices and overall software quality these days is vastly better than it was 10 or 20 years ago. Language features, new languages and mature frameworks help. IDEs help. CI setups and better automated testing helps... there is no step in the process that doesnâ(TM)t lend itself to better quality. More industry/societal knowledge helps. Weâ(TM)re just better at this than we used to be.
The big cloud flare outage wasnâ(TM)t even a
Re: (Score:3)
Re: (Score:2)
Glib, uninformed, and factually wrong edgelord bullshit. Coding practices and overall software quality these days is vastly better than it was 10 or 20 years ago.
That sounds like complete nonsense to me. Software, on average, seems to get worse, despite the better tooling available.
Re: (Score:2)
Instead of letting talented people who take time to think it through but provide superior code they only let in people that pass the latest fad of the month coding challenges. This is leading to an industry wide decline in code quality. Wonder why so many services went down this week and twitter allowed itself to be hacked badly? You have your answer.
Twitter being hacked isn't the problem with Twitter.
It's the sheer reliance an entire financial industry and society has now placed on a service that quite literally established itself with I'm-taking-a-shit-now tweets, mocking just how fucking pointless the entire service really is.
And now, in true societal FUBAR fashion, this is now a service that can now affect millions in stock prices with a single fucked-up (or hacked) tweet.
Twitter being hacked isn't the problem, but we now deserve the real problems T
Re: (Score:2)
Not just twitter. The stock market is best modeled as a roomfull of chickens where periodically someone throws a kernel of corn in or yells BOO.
That's the problem. Twitter is just one of many conduits for the bits of questionable information that the marked disturbs itself over.
Re: (Score:2)
This. We have quite enough software that evidently was ground out with as little thought as possible.
It's a double-edged sword (Score:2)
Of course big companies get a lot of applicants and many of them are equally good. These interviews are often not meant to be inviting, but indeed are the opposite. When HR has to deal with hundreds or even thousands of applications for just a couple of jobs then these tests are as much tests for character as well as skill.
This practise helps to find outstanding applicants, but it also has the tendency to support a work environment of anti-social behaviour simply because a company fills it jobs with a lot o
Re: (Score:2)
When HR has to deal with hundreds or even thousands of applications for just a couple of jobs...
This practice began with the big Silicon Valley tech firms, but then followed their former employees to all of the many small companies throughout the land, seeding a practice that might have had a justification under the situation you describe (but I question even that) was is nothing but detrimental everywhere else. As you lucidly explain the practice has toxic (i.e. bad work environment) effects even in those big firms. Their staff pays a price.
Re: (Score:2)
Of course big companies get a lot of applicants and many of them are equally good. These interviews are often not meant to be inviting, but indeed are the opposite. When HR has to deal with hundreds or even thousands of applications for just a couple of jobs then these tests are as much tests for character as well as skill.
And yet those same big companies keep claiming they can't find qualified applicants. Perhaps the problem is that by making the interview process un-inviting, they only get applicants who are sufficiently desperate. Well qualified potential applicants are less likely to be sufficiently desperate.
Whiteboard is bad interviewer crutch (Score:2)
Most interviewers in tech are other tech people with limited social skills so the white boarding allows them to kill a good part of the time HR assigned them to talk 1:1 with another human being.
It also gives opportunity to make interviewer feel smarter than candidate when they scribble some minor thing incorrectly or the entire question is a giant gotcha.
I do ask a whiteboard question but it's super trivial and can be written in a few lines in any language. It serves as the jumping off point for further d
Re: (Score:2)
Well said!
Does the interviewer insist on filming you? (Score:2)
If yes then no.
Whiteboards on Interviews? No way. (Score:2)
I've got to say, that I really don't use whiteboards to determine if somebody can code or not, and quite frankly I don't care all that much as long as they have at least a minimal grasp of the language's syntax that they claim to know.
That's not to say I don't ask coding questions, I do, but it's more about learning how a candidate thinks, over how well they can perform the nitty-gritty details of writing code. Do they understand how to take a complex problem and break it into smaller chunks? Can they ex
Re: (Score:2)
The point of the question is that I want to know, are they a loner one-man team, or are they going to seek help and escalate the problem when they cannot solve it. Where do they start talking about telling their manager their task is at risk?
This is stupid, because the proper answer is "what the programmer should do at this point depends on how the management in this company operates" with a side of "am I supposed to assume that management is competent?" Obviously you're not going to be telling a programmer at an interview "we have incompetent management", so what you're probably really testing is "is the management in the programmer's last company enough like the management in this company that the best thing to do there is the best thing to
Whiteboarding For the Right Reasons (Score:2)
Been there done that (Score:2)
A few years back I had an interview at a certain tech company that sounds like "Oogle". I was asked how to sample an unknown amount of data. I explained various ways I could select data as it came in and the trade offs of each method. One would skew towards the initial data and wouldn't be representative of the entire dataset but improve as it gathered more data. Another would be random, another would be like every nth value.
After I explained the pros and cons of each he refused to give any input on what
Re: (Score:2)
Re: Nothing of value was lost. (Score:2)
Re: (Score:2)
Re: (Score:2)
I had an identical experience at Oogle. I spent an hour developing a solution based on the guidance the interviewer had provided at the beginning, only to discover that they really wanted something different, not disclosed until the end of the session.
Re: (Score:2)
Sounds legit, but probably not what is intended (Score:4, Interesting)
In my experience when coding, what you need lon a team is a couple of people who will sit back, tap on the keyboard, and at some point throw out fantastic, readable and well-documented code, plus 1 or 2 people who can dive in when the sh!t hits the fan to bash out a proof of concept for that surprise board meeting in 30 minutes or who can hack together a temporary workaround for a critical issue that has just taken down your entire environmen (that code can then be refactored later into a more maintainable form, when everyones' stress level goes down from 110%).
If you have someone that seems to be a brilliant coder interviewing, but who freezes during that whiteboard problem solving session, you know they may not be the bet for that hacker position, but they may be the one you want writing the code that gets commited as your stable release.
Besides, everyone has an off day and a potential brainfart at the wrong moment, and it often happens when you put a coder in a meeting room with several people, in front of a whiteboard and with a marker in hand. If they were so comfortable in that type of situation that they can skate on through, you might want them for a sales/pre-sales technical consulting role instead, who then preps a mockup and framework code.
Experience (Score:2)
why does it have to be a whiteboard? (Score:2)
fuckin racist
Re: (Score:2)
Nope nope nope (Score:2)
First of all, I don't suffer from stress, I'm more of a carrier. :)
But seriously, I won't work at any place that tries to play these bullshit games with me.
If I sense that a potential employer is screwing around then I either end the interview, or in one case I just gathered my stuff, got up, and walked out. If they "lobby" me, I give them X number of minutes and if they don't show then I get up and walk out. You waste my time, I waste yours.
I'm old enough and have enough experience and money in the bank th
Re: (Score:2)
Bingo.
I like to put them in their place by right off the bat and set the tone. I tell them that "I'm at a point in my career where I generally only work on things I find interesting", and ask them what's attractive about this job, why I should come to work for them? What's compelling about their company or this position?
That's when they realize I'm interviewing them, not the other way around.
I also ask "what happened" to the last person in this position, why are they gone? Listen very, very carefully to wha
Who keeps facts in their brain anymore? (Score:2)
Is this just another excuse... (Score:2)
...to mask the real problem?
Being a brilliant coder and an introvert? Yeah, that's such an longstanding stereotype/assumption/expectation that its a meme at this point.
Being a shitty coder and an introvert? Yeah, that's a problem, because you're using the introvert excuse to mask the actual problem of being a shitty coder.
Sorry about your anxiety. Now grab this marker and get on the whiteboard and remind me again why we interview.
Re: (Score:2)
Re: (Score:3)
So what's the real problem? That companies don't realize that they are letting go of the smartest people because they value someone who can con them (ie talk well in front of them) instead of actually knowing their stuff.
I get you're hiring a coder to code. But in this day and age of "Learn to Code" we now have to weed through the candidates who actually know what they're doing vs. those who claim they do, which unfortunately is now a lot.
The real problem? More and more shitty coders pumping out more and more shitty code, which is ultimately what we're trying to avoid in the big picture. If I'm hiring you, you better be able to prove your worth as a coder one way or another. I don't mind proving that with alternative m
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
Thick skin, thin skin - neither of those have any bearing on ability, and yet you claim you want to test real ability?
You obviously didn't think it through.
slashdot (Score:2)
The study has unreasonable expectations (Score:2)
Participants in our study also reported additional stress from ...having to recall the syntax of the language.
Whiteboard coding should use pseudo-code. Compilers and text editors handle syntax for us. The only exception is if you are specifically asking about structure, like how to derive from a base class in a specific language.
Moreover, conscientious as well as sad software engineers performed worse.
What the heck are "sad software engineers?" Is this a proofreading issue?
...found that the four interviews were enough to predict whether someone should be hired at Google with 86% confidence.
Four interviews is very unusual. Most candidates would not tolerate four interviews except for places like Google or NASA. But for most companies, 1 is the norm, sometimes 2.
Their test question asked the programm
Re: (Score:2)
Four interviews is very unusual. Most candidates would not tolerate four interviews except for places like Google or NASA. But for most companies, 1 is the norm, sometimes 2.
My company, an mid-size IT consulting firm, can do 3 or more depending on position. They are pretty picky about certain soft skills since we are a consulting company. Being technical often isn't all that's required. So they will do one with the person who will be their boss, one with a senior salesperson or architect, and usually one or two group interviews with 4 or 5 people from different departments that will be working with the candidate.
Re: (Score:2)
Hazing (Score:2)
To me it looks like and feels like hazing, the kind of hazing I might expect from joining a college greek group. If you need to haze and punish me to prove my loyalty and worth then I don't want the job. If you are putting me in a situation that is normal for your work environment and I'm stressed out about it then I don't want that job. These processes are setup to exclude more than include. Excluded introverts at your own peril.
Re: (Score:2)
I can't code ... (Score:3)
More than coding (Score:2)
A company I used to work at had a very defined interview system for coders. The whiteboard part was done with the following in mind:
* Check for basic practical knowledge in the area applying for
* How does the applicant think through a problem
* How does the applicant test their solution -- just shout "DONE" or try various inputs
* Depending on their solution (once the pressure is off) can we talk through ways to optimize it - like an implementation of bitcount
* How does the applicant take suggestions / will
Yes and no (Score:2)
I want to see if somebody can think on their feet, if they can communicate, because they're useless if they can't. No matter what else they can do.
In one interview I was asked to write code to generate prime numbers. I clarified the specifications (is memory an issue?) then wrote PASCAL-ish pseudocode for the Sieve of Eratosthenes. I walked the interviewer through it to show that my code was in fact correct and analyzed its runtime behaviour. Not the most rigorous analysis, but an arm-wavy upper bound is
Weirdest interview I have had (Score:3)
I'm usually kind of a libertarian when it comes to jobs. If I didn't get the job, I assume someone else was a better fit. I have one time where I was perfect for the job. Everything about it had my name written all over it. The interview was bizarre. The actual job was a step above an entry level job I could have done in my sleep (I just wanted to get my foot in the door at this company). They bring in a panel of people to interview me and I swear it was like they were interviewing me to be vice president of the company. Never been through anything like it before or since. I didn't sleep well the night before the interview because I was nervous. I'm not sure I slept at all. I totally flubbed it.
The bottom line, from an email I got from the company: I was the most qualified person to apply for the job. I needed to work on my interview skills.
%@%#$%#$%
To this day, that is one of the few jobs I'm still pissed off about that I didn't get. And the only reason I didn't get it was because of the bizarre interview.
Betteridge says no (Score:4, Insightful)
I've conducted dozens of whiteboard interviews. You can learn a lot about someone's competence really quickly, and it has nothing to do with anxiety or pressure. A lot of people just have no clue what they're doing. They don't even know how to approach the problem, and when I finally explain it to them they can't understand the explanation because they're missing too many basic concepts.
I'm sure some of those people walked out thinking, "That wasn't a fair interview. I was just nervous." Anyone who thought that further demonstrated their lack of competence, because they totally failed to grasp what I was looking for. I've seen nervous competent people, and I've seen relaxed incompetent ones. They don't look anything alike.
Re: (Score:2)
Sometimes it's a coding test. Sometimes it's a design test. Sometimes it's a social anxiety test. Sometimes it's a test of quick thinking on your feet. Sometimes it's a test to see if you'll jump through their hoops. 100% of the time it's a test that they need to tick a box.
I've done whiteboard tests and interviews with grads. The whiteboard was about knowing basic concepts, drawing a diagram and explaining themselves without choking, or if they choked that they could recover well. The number of people that
Re: (Score:3)
Sometimes it's a coding test. Sometimes it's a design test. Sometimes it's a social anxiety test. Sometimes it's a test of quick thinking on your feet. Sometimes it's a test to see if you'll jump through their hoops.
Sometimes it's a test to see if the candidate will put up with their armchair-psychiatrist horseshit. I always fail that test.
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
I really just want to see their mind work.
Why? Are you qualified to extrapolate from what you see to knowing how their mind works, and then how that extrapolates to what happens on the job? Have you conducted studies on your failure rate, both in the people you hire, and the people you reject who go do an excellent job elsewhere?
I really just want to see how they go about problem solving.
Again, why? How does knowing the detail about how they go problem solving help? People solve things in different ways developed to their particular experience. As long as they are actually capable of solving the problem wit