6 ms·

One thing you're testing for is how well their brain works with an excess of adrenaline floating around in their body. All of the best interviewers I've known t

by trotsky 14y ago

One thing you're testing for is how well their brain works with an excess of adrenaline floating around in their body. All of the best interviewers I've known take great pains to make the interviewer feel comfortable and chatty - after all, they aren't going to be filled with interview anxiety when they're actually working there. A minority of your applicants are going to perform badly in that situation because the adrenal glands will effectively reduce the output prefrontal cortex below an acceptable level to solve a problem that they'd be able to solve handily in a situation that didn't involve so much performance anxiety with strangers. I doubt that constitutes a majority of the people you pass on, but it's definitely going to be double digits. I've never hired using a white board coding test, and out of 30 or 40 folks i've had to let 2 go pretty quickly because it was clear they couldn't do the work, but the rest did as well as somebody who would have passed one of your quizes - I'd expect we'd see similar long term failure rates in both sets. I'm sure it makes sense at google where the hiring procedures don't leave much room for error, but I'm sure I've been able to hire some great folks that you would have missed. Arguably the best heads down brilliant engineer I've hired (well tied for #1) completely melted down on his first interview. Tried again with drinks and a show at Yoshi's and he did so well I would have hired him to be my boss. He easily pulled 3x his own weight for us and was very loyal and easy to work with. So many people go on and on about how hard it is to hire yet they go around systematically rejecting all the same candidates for all the same reasons.

davidkatz 14y ago
"A minority of your applicants are going to perform badly in that situation because the adrenal glands will effectively reduce the output prefrontal cortex below an acceptable level" I have an off topic question. I don't see what the problem being discussed has to do with how glands reduce activity in certain parts of the brain. Sounds to me like a fancy way of saying 'people don't perform well under stress'. You could of course ask - "what mechanisms control human behaviour, and how does stress specifically affect body systems?" Then, great, let's hear all about adrenal glands and prefrontal cortexes. When that's clearly not what the question was though, I find it a form of poor communication. Anyone have any thoughts on this?
- phaus 14y ago
  My thought is that it's a real phenomenon that occurs in a certain subset of people during a job interview, therefore, its obviously relevant
  - Arelius 14y ago
    I feel the problem is that, few of us actually being neuro-scientists; While saying something like "people don't perform well under stress" is something that is close enough to common knowledge, instead saying "adrenal glands will effectively reduce the output prefrontal cortex" leaves the reader with a whole bunch of unanswered questions: * Does a stressful interview necessarily actually increase adrenaline? * Is there actually a link between incresed adrenaline and reduced prefrontal cortex output? * What role does the prefrontal cortex play in answering interview questions? And while these may all have good answers, it just throws up a whole bunch of mental '[citation needed]' annotations all over the whole argument, that can just be skipped entirely if we aren't talking about neuro-science. And if we insist on talking about the bio-chemestry, we should probably go into the details.
  - davidkatz 14y ago
    Lots of real phenomena occur during an interview. For example, you could go one level deeper and describe the molecular mechanisms between adrenaline and whatever 'receives' adrenaline. That would sound something like: "What happens in job interview is that adrenaline binds to a variety of adrenergic receptors, which triggers a number of metabolic changes" [1] It'd be just as real, also clearly connected, but also quite clearly not what the conversation is about. [1] http://en.wikipedia.org/wiki/Epinephrine http://en.wikipedia.org/wiki/Epinephrine
- monksy 14y ago
  Add major jet lag to that and you'd have one of my previous interviews.
- cwp 14y ago
  I think it's a little more specific than "people don't perform well under stress." A software development job may well entail stressful situations—for example, "The founder got interviewed on Oprah and the servers are melting!" But that kind of stress is different from a interview stress. There are definitely folks that will step up to the challenge of melting servers, but go to pieces under the social stress of an interview. Now, maybe getting into the biological mechanism by which they go to pieces is too specific, and trotsky could have communicated more effectively by eliding it. But it's an important point, and I don't think it reduces to just "people don't perform well under stress."
rgovind 14y ago
How did you happen to call him for second interview if he failed first one? Were the two interviews seperated by 6 months or so?
coldtea 14y ago
>One thing you're testing for is how well their brain works with an excess of adrenaline floating around in their body. Unless you are specifically hiring people for a death march kind of project (which I find unwise anyway), targeting people good at high-stress situations doesn't seem very productive. If your programming projects are a series of high-stress, high-adrenaline runs, either you work for NASA or you're doing it wrong.
31reasons 14y ago
As a company you should always advertise if you want your employees to work under a lot of stress. I would for one never choose to work for such groups.

by trotsky 14y ago

davidkatz 14y ago

"A minority of your applicants are going to perform badly in that situation because the adrenal glands will effectively reduce the output prefrontal cortex below an acceptable level" I have an off topic question. I don't see what the problem being discussed has to do with how glands reduce activity in certain parts of the brain. Sounds to me like a fancy way of saying 'people don't perform well under stress'. You could of course ask - "what mechanisms control human behaviour, and how does stress specifically affect body systems?" Then, great, let's hear all about adrenal glands and prefrontal cortexes. When that's clearly not what the question was though, I find it a form of poor communication. Anyone have any thoughts on this?

phaus 14y ago
My thought is that it's a real phenomenon that occurs in a certain subset of people during a job interview, therefore, its obviously relevant
- Arelius 14y ago
  I feel the problem is that, few of us actually being neuro-scientists; While saying something like "people don't perform well under stress" is something that is close enough to common knowledge, instead saying "adrenal glands will effectively reduce the output prefrontal cortex" leaves the reader with a whole bunch of unanswered questions: * Does a stressful interview necessarily actually increase adrenaline? * Is there actually a link between incresed adrenaline and reduced prefrontal cortex output? * What role does the prefrontal cortex play in answering interview questions? And while these may all have good answers, it just throws up a whole bunch of mental '[citation needed]' annotations all over the whole argument, that can just be skipped entirely if we aren't talking about neuro-science. And if we insist on talking about the bio-chemestry, we should probably go into the details.
- davidkatz 14y ago
  Lots of real phenomena occur during an interview. For example, you could go one level deeper and describe the molecular mechanisms between adrenaline and whatever 'receives' adrenaline. That would sound something like: "What happens in job interview is that adrenaline binds to a variety of adrenergic receptors, which triggers a number of metabolic changes" [1] It'd be just as real, also clearly connected, but also quite clearly not what the conversation is about. [1] http://en.wikipedia.org/wiki/Epinephrine http://en.wikipedia.org/wiki/Epinephrine
monksy 14y ago
Add major jet lag to that and you'd have one of my previous interviews.
cwp 14y ago
I think it's a little more specific than "people don't perform well under stress." A software development job may well entail stressful situations—for example, "The founder got interviewed on Oprah and the servers are melting!" But that kind of stress is different from a interview stress. There are definitely folks that will step up to the challenge of melting servers, but go to pieces under the social stress of an interview. Now, maybe getting into the biological mechanism by which they go to pieces is too specific, and trotsky could have communicated more effectively by eliding it. But it's an important point, and I don't think it reduces to just "people don't perform well under stress."

rgovind 14y ago

How did you happen to call him for second interview if he failed first one? Were the two interviews seperated by 6 months or so?

coldtea 14y ago

>One thing you're testing for is how well their brain works with an excess of adrenaline floating around in their body. Unless you are specifically hiring people for a death march kind of project (which I find unwise anyway), targeting people good at high-stress situations doesn't seem very productive. If your programming projects are a series of high-stress, high-adrenaline runs, either you work for NASA or you're doing it wrong.

31reasons 14y ago

As a company you should always advertise if you want your employees to work under a lot of stress. I would for one never choose to work for such groups.