[wglb]

There is a lot of value in the comments on this article, but I am afraid that the idea that you can usefully hire someone based on the answer to one question is a bit oversimplified. In any project that I have done that spans any length of time, often the first code I wrote doesn't look up to snuff with what I wrote most recently. The later code looks better by virtue of a better feel for the environment and the tools and some obscure abbreviation-finder that seems to be at work in the background.

One concern that I would have if you think the worst code that you have seen is stuff you wrote, perhaps you are not reading enough of other people's code. I bet there is a lot of that amongst us all.

In the vein of simple questions that help in an interview, particularly for someone who purports to be an expert, is "Tell me five ways that it won't work". An expert ought to have some scars, and have learned from them, no?