Yifan Sun recently graduated from UCLA, after 5 years of pursuing her PhD in Electrical Engineering. Today, she is a Researcher at Technicolor, working at the intersection of machine learning, optimization, and algorithm design. In a recent exchange, Yifan shared these thoughts with us.
Advice for Oliners seeking to attend grad school?
Read more papers. Research happens at breakneck speed, and the only way to keep track of a) what you’re interested in and b) what you can make progress in is to constantly be aware of what other people are doing. I’d say this is still my greatest weakness.
Slow down. If I were to compare my time at UCLA and Olin, the biggest difference is that quality >> deadlines. Negotiate for fewer deadlines so that you can have better quality work.
Do you feel like what you are doing differs from what most Olin students do after graduation?
Now that I’m in the Bay Area, I interact with a lot of more business-y people, as well as the usual nest of researchers. As a researcher, I have the luxury to spend more time doing and less time selling, but I’m sure a business person would say something different. Overall they’re both very challenging, rewarding, and impactful jobs.
Tell us more about your experience at UCLA.
UCLA is very different from Olin ;) The school itself is like a small city (within a huge city) and because of that, you can try a lot of different things with basically no consequences. (Who’s going to remember?) Probably the biggest culture shock in going from a small school to a big school is how much the burden shifts to you, the student, in order to get anything done. That sometimes means stalking your adviser for a meeting, or cold calling every financial office in the UC system to figure out why your paycheck hasn’t come through for months, or finding your own outlets to promote your projects. Academia in particular can be a bit snobby, so it’s extra important to pick goals that fit you, and not blindly listen to advice from people (including post-grad bloggers like this one).
UCLA also has a much more traditional teaching philosophy than Olin; students give feedback but they don’t shape a course, and everything is starkingly formulaic: 10 weeks of lecture, 1 midterm, 1 final, sprinkled homeworks. It sounds like your worst nightmare but it’s also shockingly effective, especially if you don’t really have the time to give 100% of yourself to that course.
What is a typical day like for you?
Every day is different, but near the end of UCLA my routine was something like: spend the morning avoiding trucks in the bike lane, then sit through some really theoretical class taught by an elderly prof who then gives you the tex code for the problem set (because who uses pen and paper in today’s world?), have a heated debate over lunch about politics and society, spend the afternoon reading or coding, then end the day hanging in Westwood Village with other night owls.
After I joined Technicolor my days were more structured, but the free-form part of self-driven research is more or less persistent.
What do you feel you are doing that's innovative?
The job of a STEM researcher is to push the boundary of science/technology and produce things that are truly new. I personally think the story is more like this: researchers are people who get a ton of gratification out of learning something that, the first time around, really made our brains hurt. We became masochistically addicted to the challenge of de-tangling our brains, and the only thing we love more is talking to other researchers about the latest brain tangle. Deep down we don’t actually care if it’s innovative or impactful, but people tell us that if you keep this community alive and vibrant and herd the topics toward the impactful stuff, then one day innovative and impactful stuff will emerge, so we sleep soundly at night.
Do you feel that you have made a difference in the world since leaving Olin?
I’d like to think that by pursuing what I like, I will inherently make choices that make me a positive contribution to my community. </cheesy>
How did your time at Olin prepare you for the real world?
I think at Olin we are trained to see everything from a high level view. “What is the objective of the project?” “Who are the major players?” “What are my obstacles and deadlines?” That, and a ton of programming practice are probably what served me the most. (Seriously, no one teaches you programming in the real world.)
Probably the stuff I wish I had more was practice honing more theory-based skills, like reading papers, taking more standard math classes, reading papers, forming and following through with self-driven projects, reading papers, etc.
Do you have time for fun? What do you like to do?
Grad school was busy but there’s a ton of “me time” too. At UCLA, I did a lot of biking, I had a pet rat, jogged at the beach, road trips, etc. Now that I’m working I have a lot less time, but I try to maintain something (piano lessons, occasional game nights, etc.) I’m a believer in the work/life balance!
Advice for current students?
If I were to redo Olin, I’d spend a lot less time cooped up in my room being nerdy and a lot more time trying things (not nerdy stuff!) I’ve never tried before. (So typical Olin spirit, but more.)
I’m also on the bandwagon of alumni who wished that Olin had prepared me more for the traditional mathy/sciency subjects. It’s true that a school can never prepare you for everything. One thing I really liked about UCLA’s grad program is that you pick 7 classes -- any classes -- and they’re hard as hell but you do them, and that’s 7 subjects you just became an expert in. I think that’s a balanced way of doing math; pick one subject per year and learn it the traditional way, without complaining about “how it’s taught” or “how I don’t learn this way”. Learning how to learn in different ways is just as important as anything else. And, if you decide not to be an expert in something, taking a course is the best way of being “good enough to scrape by”. But most importantly, carpe your diem.