I don't believe that. I'd really like a 21st century study showing that devs with > 3 years of experience fail fizz buzz, and that failure is predictive of being a bad developer (and not being a shitty interviewer)

Yes, of course a dev with > 3 years experience should be able to fizz-buzz in their sleep! That's the point!

It's not devs with > 3 years experience or even > 6 months. failing fizzbuzz. It's people _pretending_ to be devs with > 3 years experience. Or folks who started in programming years ago and drifted so far they can't get back to nuts and bolts. Or people who know the principles but can't express them. Or any number of other cases.

Like other folks have posted here in response, I've seen some really sad fizz buzzes in my day, often from people who had no business being near a computer let alone interviewing for a tech job, but somehow landed a phone screen.

And let's never forget the best/worst fizz-buzz ever: http://thedailywtf.com/articles/The-Fizz-Buzz-from-Outer-Spa...

Bear in mind that everyone here on HN is here because they're interested to learn about advancements in technology (and matters of broader interest) - it's too easy to mistake this little echo chamber for the entirety of the talent market.

I often read people say they'd be "insulted" if asked to do fizz-buzz, or would "walk out". I think that's an over-reaction - fair enough they need to make sure you're not useless and lying on your CV. As long as there's a more interesting follow-up question, I don't see the harm.

If you don't believe Jeff Atwood, I'm not sure why you'd believe me based just on my anecdata, but at a previous job I gave a simple filtering programming task to interviewees (not exactly fizz buzz, but very close. Single loop, multiple 10-line solutions possible).

Out of 10, only 1 solved it well, 2-3 solved it acceptably (only edge case bugs), the rest never solved it, or had obvious bugs or code that would never pass peer review.

Of course, I could just be a shitty interviewer, but I always had a (changing) peer present during interviews, and I hope we were an honest enough company that they'd tell me if they thought I was.

Yes, this is typical.

In 2013, I interviewed someone who had allegedly been working for a few years as a support engineer writing Ruby (using the "watir" gem) who could not even write the loop part of Fizzbuzz. They went completely silent on the call when I pasted the Fizzbuzz requirements into a shared code editor we were both viewing. After a few moments and my gentle prodding, they admitted they didn't even know how to begin writing the loop, so I thanked them for their time, wished them luck, and ended the call.

You don't have to believe it - check out the comments in Jeff Atwood's post, where many people assert they can't believe people can't solve FizzBuzz, look they've solved it right here, and post their incorrect solutions.

I've never given a fizzbuzz test, but I can assure you that many people fail to write a simple for... loop to traverse an mxn square. I've seen it so many times that I have absolutely no problem believing that the same people would fail fizzbuzz.

You really don't realize how bad candidates can be until after you've interviewed dozens of them.

One of my teammates had a phone screen with a guy who didn't know what prime numbers were...

You can't blame them for not keeping up with the latest Amazon offerings.