by Lori Ann White
In the series, "Where are they now?" we check in with KIPAC alumni: where they are now, how they've fared since their days exploring particle astrophysics and cosmology at the Institute, and how their KIPAC experiences have shaped their journeys.
Next up is Yvonne Edmonds, who spent her time at KIPAC searching for signs of dark matter in the data gathered by the Fermi Gamma-ray Space Telescope (FGST). She's now putting her skill at data analysis to work in the realm of healthcare and pharmaceuticals.
LW: Can you tell us a little bit about your academic background before you came to KIPAC?
YE: I got my BS in physics in Baltimore, MD, at the University of Maryland, Baltimore County. I tried a lot of engineering first, so I was a little late to physics.
LW: What was it about physics that drew you?
YE: Ultimately it's the simplicity—I think physics is really elegant. I like thinking that I can understand just a few things and use that knowledge to learn more about the universe.
LW: Did you do any research while you were an undergraduate?
YE: I got to work with one professor while I was an undergrad on AIRS—the Atmospheric Infrared Sounder—which is an instrument on a NASA satellite called Aqua. It's one of their earth science satellites. I also spent a summer at UC San Diego on an REU (Research Experience for Undergraduates).
LW: What did you do for AIRS?
YE: I was quantifying the amount of sulfur dioxide in the atmosphere and creating a graphic interface that made it easier to visualize on a larger scale. Then I graduated in 2004 and started at Stanford in the fall of 2004.
LW: Why did you choose Stanford and KIPAC?
YE: Well, my mom is from Berkeley originally and I still have family in the area. One uncle has an engineering degree from here. Stanford itself is really well-rounded, with excellent academics, great research opportunities, and lots of extracurricular activities. I also knew about the rotation system, where graduate students get to try out research in different areas before deciding what they want to work on, and I didn't know what physics I wanted to do. So that was a selling point. I did rotations in condensed matter, biochemistry, and electrical engineering, but the benchwork wasn't much fun. Me and the lab wasn't the best fit. (Laughs.)
LW: What do you like?
YE: I like the really basic questions. So cosmology interested me. Analysis and data analysis, computers and programming. I enjoyed analysis more than wet work where I was touching things. Maybe I just like telling the computer to do something. (Laughs.)
LW: There are lots of basic questions to investigate at KIPAC. What did you choose? 
YE: I worked with [KIPAC professor] Elliott Bloom, using data from the LAT [Large Area Telescope, the main instrument on the FGST] to do a dark matter line search. We looked over the entire sky at areas where we thought there'd be high densities of dark matter.
LW: Can you explain what a line search is?
YE: According to a lot of dark matter models, dark matter particles can collide, converting to gamma rays. That could show up in our data as a spike, or line, at the energy of the annihilating particles. So there could be evidence of dark matter in the gamma-ray sky, like a smoking gun. That's why this was so interesting.
LW: What did you find?
YE: We didn't see anything indicative of dark matter, but we did set limits, which is very important. There are a lot of dark matter models out there, and they predict certain things, like the mass and how often particles interact. We can rule out models based on what we don't see—okay, this particle doesn't fit because we would have seen a certain signal but we didn't.
LW: When you graduated, did you give any thought to continuing in academia? Dark matter is still out there, just waiting to be found—
YE: No, I was pretty sure I wanted to go into industry. What's good about grad school is that you get a chance to really see what being a professor is really like and what life in academia is really like. I did want to answer basic questions, but I also wanted to make a faster impact, to see the results of what I did more quickly. But I don't regret getting a PhD in physics. It's a degree where it's really easy to reinvent yourself because of the skills you acquire. You know how to problem-solve. If you don't know what you want to do, physics is a good degree to have. (Laughs.)
LW: What have you been up to since KIPAC, then?
YE: I started at KLA-Tencor (semiconductor equipment) after getting recruited at a Stanford Career Fair—KLA-Tencor actually hires a lot of Stanford physics grads. It was a good opportunity to learn a lot because I came in as an engineer for products that were already in the field so I learned about customer pain points, but then I was the owner of a prototype and then I got to work on a project that was two generations out, so I was able to learn about the whole product cycle, just not on all the same product. But I also discovered that analysis really is what I like. I think it was good for me to try hands-on again but I think I really was glad to get out of the lab. (Laughs.) And I knew there was so much to learn in data science. It's such a broad field right now that it doesn't even mean the same thing to everybody. It can mean you're doing something predictive, or descriptive, or what data to get in the pipeline.
LW: How did you change direction?
YE: I found out about the Insight Fellowship. Josh [fellow KIPAC graduate Joshua Lande] did it and I just thought it looked very interesting. While I was there I worked on an application that recommended movies based on title keywords or other specific criteria. It also forecast what movie earnings based on things like the plot and the cast—kind of like fantasy football for films.
LW: That sounds like fun.
YE: It was. After that I went to McKesson, which is a pharmaceutical company, but we don't make drugs. We buy them from suppliers using economies of scales, and sell them to pharmacies. I work on questions like, can we better manage our supply of certain drugs by understanding when people will demand them. Flu is a good, if simple, example. Everyone knows when flu season is, but if clinics and pharmacies don't get the vaccine in at the right time there's a big problem. Pricing, locations, throughput—what we do maps to most other supply chains. Another reason I was interested in this was that they have a lot of data but not a lot of people are doing predictive analysis so there's a lot of work to do—there's a lot of new stuff to develop. Plus I'm gaining a lot of the experience of working in a business.
I pick my jobs based on the new skill sets I want to learn and the domain I want to be in, and this a totally different way of thinking that I think is good for me to experience—it's not really even money, or the profit motive, it's just being responsible and accountable and able to deal with pressure and deadlines.
LW: You sound like a budding entrepreneur.
YE: (Laughs.) I wouldn't go that far.
LW: Do you have any favorite memories of KIPAC or advice to other graduate students?
YE: I really enjoyed the graduate student seminar program (called SASS). Graduate students from all over the lab at SLAC would get together and present their research, and no professors allowed. (Laughs.) It was a great way to meet all the physics students, not just the just the KIPAC students. Which reminds me—another great thing was the networking. I didn't know then how important it was but now I'm really glad for all the physics students I got to know. So there's one piece of advice. Make sure you get to know everyone. That's something they teach you in business, but in physics you might just think, "I'm going to work on my thing. Why do I need to know someone who works on that thing?" But that's not enough for a lot of reasons. You may end up working on that thing, or it may give you insight into your own thing. The broader your overall knowledge, the better-equipped you'll be to do your own thing and be successful at it.
Related reading
Fermi LAT Search for Dark Matter in Gamma-ray Lines and the Inclusive Photon Spectrum (May 2012) (All plots in this paper were created by Edmonds)
