[eschutte2]

Algo/puzzle/riddle is not the same as whiteboard. I'm all for getting rid of the former, but I have no problem getting up and writing on a whiteboard as long as it's used for its correct purpose, viz. as a tool for talking through a problem together.

I once had an interviewer give me a somewhat interesting puzzle to work through on the whiteboard, but then he got all stammery and flustered when I asked a few simple follow-up questions about it. I think there are people out there who just can't handle humans and have to use the whiteboard as a shield.

[donw]

While problem-solving is fine, I am not a fan of algorithms questions.

Not because algorithms are unimportant, but because being able to code a heapsort from memory is, from a skills perspective, effectively useless.

As the creator of Homebrew put it: "90% of our engineers use the software you wrote, but you can’t invert a binary tree on a whiteboard so fuck off."

To me, it is enough to know what algorithms you could bring to bear on a given problem, and have a rough knowledge of their complexity.

The things I care more about are: Can you tackle a messy-real world problem, one that doesn't have a clear solution? Do you collaborate effectively as a member of a team? What about your software engineering skills -- is your code well-tested and readable? That sort of thing.

That latter point is important. The best engineer in the world is a business liability if nobody else can maintain their code.

[luca_ing]

> Algo/puzzle/riddle is not the same as whiteboard. I'm all for getting rid of the former, but I have no problem getting up and writing on a whiteboard as long as it's used for its correct purpose, viz. as a tool for talking through a problem together.

I agree.

We did a few interviews last year, and we had a few whiteboard-ish (people could do it in their heads if they preferred) questions.

However, the whole purpose of these questions (and this was made explicit to the candidate) was to serve as an anchor for a technical discussion, and as a tool to observe them reason about a problem. And as a way to observe them asking questions, because we were intentionally incomplete (and candidates knew).

I felt it worked very well for us.

Also, sometimes such an interview would just be plain fun for all involved. We took it as a strong signal if that happened.