I had a phone interview with D.E. Shaw today (I’m applying for a summer internship there). I was kind of afraid I’d get asked about what I wanted to get out of it and where I wanted to be in ten years, which I would’ve had an awkward time answering (that is, arguably, a bad thing in its own right, but never mind). It turned out to be basically fact-based, so that was okay. I talked about the work I did with an astronomy professor last term; the interviewer had me talk about how we processed the data, then asked me some questions about the statistics of it. I remembered the process pretty well, though I guess I wasn’t very clear and had to retry some of it. I remembered the Poisson distribution, but he asked about the conditions for one to be approximated by a normal distribution, which I didn’t really remember. I said it’s acceptable when the expected number of events is at least 20, which turns out to be about right.

After that he asked how to do quickselect, which I answered fairly well, and how to partition an array in place, which was okay, though I didn’t say it very smoothly. Then he asked a pretty simple probability question, which I sort of figured out how to do pretty quickly in kind of a neat way, but it was different from what he was expecting and I didn’t explain it very well, so that kind of fell down. I got the expected method with some prompting, but I really should’ve made it more clear that I actually knew what I was doing. I think he did somewhat recognize what I was saying, at least. Still, I think that was quite a trip-up.

“I will stomp on things to focus my mental energies, or ‘menergies’.”
– T-Rex