Schaeffer Leverages Cube Experience into High-powered Position

Ben Schaeffer got his Ph.D. from the University of Illinois and some valuable real-world experience at the Beckman Institute before leaving his Midwestern roots behind for the opportunity and bright lights of Wall Street and New York City. Schaeffer left a huge mark at Beckman as the primary author of the software code that powers the Cube, the immersive virtual reality environment operated by the Institute's Integrated Systems Laboratory (ISL). Schaeffer now works on Wall Street as a quantitative analyst in the field of computerized securities, but he hasn't forgotten his alma mater or the people who helped shape his experience at Beckman and Illinois. The Office of External Relations recently asked Schaeffer about his experiences at Beckman and Illinois, his successful career, and other topics in an e-mail question-and-answer session.

First, could you tell us about your personal background, where you grew up and went to high school, your degrees, and how you ended up at Illinois?

I grew up and went to high school in Evansville, Indiana. My degrees are B.S. in Mathematics from the University of Chicago and Ph.D. in Mathematics from the University of Illinois at Urbana-Champaign. My thesis advisor was Carl Jockusch, and I studied mathematical logic.

I was told that you alone wrote the code (Syzygy) that powers the Cube and that can also be used for other platforms for PC and PC-cluster-based virtual reality environments and other graphical applications. I was also told that it was unusual for someone to do both the math and the code for software like that, so it was a big accomplishment. Is that something you're most proud of from your time at Beckman and Illinois? Did it aid your future career, either getting the job you have or in writing other codes? If you wanted to say more in depth about the process that resulted in Syzygy, that would be great also.

As a teenager, I'd done a lot of computer programming. At the end of the Ph.D., I got interested in doing something practical (as opposed to purely theoretical) and decided that working with computers might be a good skillset to develop. One of my big hobbies was painting, so doing something artistic appealed to me as well, and it turned out that the Beckman (via NCSA) was a big center for Virtual Reality (VR) at the time, with (Beckman faculty member) George Francis teaching a class in mathematical visualization that stressed VR. This was a unique opportunity and I pursued it, taking George's class, along with a number of first year computer science (CS) grad classes while writing my thesis. After graduation, I worked for Daniel Reed in his Pablo Group and got valuable exposure to the supercomputing world. In that context, I saw some of the preliminary work people were doing with scalable display walls and realized that a PC cluster architecture made sense for CAVE environments as well. The potential cost savings were incredible, and, indeed, the first visualization cluster for the "Cube" cost about $50K, compared to the $1.5 million for a SGI Onyx2. You don't often see a factor of 30 cheaper!

My involvement with Beckman began when Rachel Brady and Hank Kaczmarski hired me into the ISL to work on software for what would become the Cube. I remember sharing the "PC cluster VR" scheme with them in that first interview and discovered they were already thinking about the same thing. It really was amazingly fortuitous. At the time, I was convinced that the VR niche was potentially quite big, assuming that costs could be brought down, and thought this was a great opportunity to work on something that could really have an impact. In retrospect, it was a really wonderful experience to work on a new technology (that we weren't really sure would work as well as it did), and I'm proud to have turned out a product that served a need at the university. However, the VR niche (vis-a-vis society) has turned out to be much smaller than I thought it would be. It's pretty clear that the larger trend in computer interfaces (those used by millions of people) is smaller (think cell phones and handhelds) not bigger (like VR). Once I saw this clearly, it started to seem like time to move on.

The segue to a new career direction was a little unexpected. The main trick in writing Syzygy was writing reliable, high quality, and high performance real-time distributed systems code. Making the quality of the PC cluster display equal that of the Onyx2 (and, incidentally, making the whole thing reasonably user friendly) wasn't easy. It turns out that real-time distributed systems programmers are in high demand on Wall Street for writing computerized trading systems. So, those particular skills, combined with my math background, led me to my current job as a quantitative analyst at Lehman Brothers (a "quant" in the local slang). Regarding Syzygy, I'd say that I "invented it" and wrote the original functional system. However, Camille Goudeseune and Jim Crowell deserve substantial credit as co-authors/co-developers. They are impressive scientists/engineers and contributed substantially to the final product. I was lucky to work with them.

What other accomplishments would you like to mention from your time at Beckman/Illinois?

Hank Kaczmarski did a very good job of collaborating with artists in the university community. In particular, I enjoyed working with Yu Hasegawa-Johnson and Luc Vanier on dance performances. With Hasegawa-Johnson, we did a "long distance collaborative dance" between Champaign and Los Angeles. And with Luc, we did a performance at the Krannert Center along with an installation at the Krannert Art Museum. Later, Hank took Syzygy and built a permanent museum exhibit in the Art Museum. So, in general, I was very pleased to be involved with arts/technology cross-disciplinary work.

Could you talk about the Beckman/Illinois people who influenced you, and the ways in which they helped you; people such as perhaps George Francis, or Hank, or whoever you care to mention.

George Francis was a big influence. His visualization class helped me get back into programming after a long hiatus. Also, he became a valued collaborator later when I was with the ISL, and he used Syzygy and the ISL VR facilities for a number of his summer classes for undergrads. These are part of some sort of "research experience" campus program, I think. In any case, George devotes quite a bit of energy and talent to mentoring young people and really deserves a lot of credit for service to the University.

Hank is someone I respect tremendously and feel very fortunate to have worked with. First of all, he is a down-to-earth person and extremely loyal and hard working. Second, he is a real genius with manipulating/managing physical stuff ... electronics, displays, shipping stuff here and there, you name it. The hack he pulled off getting the Cube frame and screens into the Beckman basement is simply beyond belief. One of my favorite things regarding working at ISL was his can-do, no-excuses, hard-working attitude and I hope that rubbed off on me a little bit.

I'd like to say how much I appreciated my ISL colleagues Camille Goudeseune and Jim Crowell. They both taught me a lot about programming. I'm a much better programmer for having worked with them. Jim, being a psychologist, taught me about human perception, how to be always open to doing things in a new way, and turned me on to the Python language. Camille amazed me with his ability to create gadgets of all kinds, showed me that good error handling/error messages are critical parts of programming, and, in general, was a relentless force in writing code professionally, in the right way. The greatest thing about working at Beckman was being able to interact with top notch people such as these.

What is your job title at Lehman Brothers and how long have you been there? Could you describe as much as you are able what you do there? How did your experience at Illinois and Beckman prepare you for your current position?

I'm a quantitative analyst at Lehman Brothers (a "quant"). Been there about 2 years. I work with algorithmic trading systems (so computerized trading of securities).

Beckman prepared me for this job in a number of ways: being able to produce results under time pressure; having to create a working "production" system; being able to analyze project requirements, do a design, and carry things through to their conclusion. But, most importantly, being in academia allows one to explore new technologies and new kinds of software. Following things through to their logical conclusions (making lots of mistakes along the way and going into dead ends) is a real luxury that isn't always present in the business world. Also, the freedom and opportunity to create (be it in software, some technology, some math theorem) is precious. Taking real advantage of these opportunities gives one condensed practical experience and a real edge in any technology intensive/ intellectual capital intensive business going forward.

How did you become interested in math and computer science? Is there one area, such as working with graphics, that intrigues you more than others?

I became interested in computers via computer graphics and video games when I was 10-years-old, so, in a sense, the VR was returning to my first love. But I've always liked to do a number of different things. It sounds corny, but for a long time, I got into a search for "truth", which led naturally to mathematical logic (which is also called "the foundations of mathematics" by some) and to leave more applied pursuits (like computer programming) behind. And to have really followed that search to a conclusion (not *the* conclusion of course), being exposed to a number of the world's real experts, was a real privilege. More recently, I've been interested in probability, stochastic differential equations, and modeling, really more applied than pure mathematics, which is a real switch for me, since I'm back to being interested in math again, and active in there again, but it's a new area for me. So, in general, always learning and always moving on.

Do you live and work in Manhattan? If so, how are you enjoying living and working there compared to life on the prairie? Any great memories from your time at Illinois?

I work at the Lehman Brothers headquarters in Times Square. My wife and I live near downtown Brooklyn. The transition was a little tough for me since I've been a midwesterner almost all of my life and lived in Champaign in particular for 13 years. Champaign is low to the ground, relatively empty (not densely packed), and very quiet. NYC is the opposite. So, for the first year, I found myself listening to an iPod whenever I was out, to drown out the chaos/newness, which was a little intimidating. But, after that, I suddenly adjusted, got rid of the iPod, and enjoy paying attention to the city environment when I'm out.

Here are four wonderful memories... First, going to the comic book store on Green St. in Campustown every Wednesday to pick through the new comics. It was a constant in my life during my time in C-U. Next, watching the Illini almost win the NCAA basketball title in some downtown Champaign sports bar. Third, thinking back to my first few years in grad school, being on fellowship over the summer and just sitting around day after day studying mathematical logic and not doing too much else, the air conditioning in the shared grad student office and the distinctive "Omega series" in logic books, with their yellow-orange covers. Fourth has got to be all the mornings when I show up to work and take the Cube for a spin, just feeling the experience, in the seemingly infinite blackness of that basement room...

Do you have any advice for students or others who may want to follow a similar path to the one you followed?

Learn some useful skills. These will let you support yourself and give you opportunities. Learn some things that don't seem practical too. Otherwise, you won't have the breadth of knowledge and flexibility needed to try new things and take risks. Always make sure to build and do instead of talking about building and doing. Don't leave a job unfinished, especially the last 5% that takes the majority of the work.