In retrospect, I topped out in my upper division coursework for the math major. I still got B's and even some A's in those courses (Real Analysis, Abstract Alg

In retrospect, I topped out in my upper division coursework for the math major. I still got B's and even some A's in those courses (Real Analysis, Abstract Algebra), but I was reduced to studying solution sets rather than mastering the material creatively. This was when I got more interested in applications (optimization, scientific programming, etc). I figured I'd pick a grad program in Industrial Engineering so that I could get away from math and more into applications. Unfortunately, I picked Berkeley. A graduate intro to optimization was pretty much a series of proofs about convex sets. Same deal for stochastic processes (it might be different with different professors, though). BTW, this isn't meant as a knock on Berkeley - this theory focus is a big part of why it's such a powerhouse program. But it probably wasn't the right choice for me. I've heard Stanford is more flexible (I've also heard the grass is greener, so who really knows). At this point, I topped out in grades as well as understanding, because I didn't have the motivation to continue. I tanked, and squeaked by with a MS. I actually missed problems that I would have gotten right as an undergraduate. I was thinking "I've had enough of proofs, I just don't want to do this anymore." In a way, I got a lot out of the experience. My interest in academics waned, and I pretty much just started hacking with like-minded students. In other words, I finally, belatedly arrived at the process of becoming self-educated, which is really the only kind that sticks, right?