I've done this in two different contexts. One, when you fly somebody out there is often a fixed amount of time to interview them. People run long, etc., and yo

I've done this in two different contexts. One, when you fly somebody out there is often a fixed amount of time to interview them. People run long, etc., and you just don't get enough time to really dig into their skill set. But I liked them, and just want to finish what I started. Two, things didn't necessarily go so well in the interview, or somebody else raised an issue ("did you ask about XXX - they seemed vague on it?" "No, I didn't") that I want to follow up on. One way or another you don't have enough information to make a decision about the person, but you are still interested. If you do futz with your github I'd recommend doing it in a deliberate and proactive way. Instead of just checking in changes, maybe fork or branch it, do the work, and far more importantly, document what you did. Then tell the company you'd like to talk about this project and talk about what you have learned, and how you are working on self improvement. I don't care if somebody doesn't know something; I do care whether they have the drive to learn, because this job is often drinking from a fire hose. Also, I'd be very impressed with somebody who is thinking about things like how to best present themselves, that take initiative to try to solve my problems, where the problem in this case is how do I accurately measure this person in a short time period. If you try to do that, I'm going to want to hire you.