Mathematics at BP (feat. Chris Arettines)
Alumni Aloud Episode 74
Chris Arettines earned his PhD in Mathematics from the CUNY Graduate Center and is currently a Quantitative Developer at BP. In his career, Chris has held a few different data science positions.
In this episode of Alumni Aloud, Chris speaks about translating graduate school research skills into industry success and how to develop a successful career in the world of data.
This episode’s interview was conducted by Misty Crooks. The music is “Corporate (Success)” by Scott Holmes.
Listen to the episode below; download it; or stream it in Apple Podcasts, Google Podcasts, or your preferred podcast player.
Podcast: Play in new window | Download
Subscribe: Apple Podcasts | Google Podcasts | RSS
VOICEOVER: This is Alumni Aloud, a podcast by Graduate Center students for Graduate Center students. In each episode, we talk with the GC graduate about their career path, the ins and outs of their current position, and the career advice they have for students. This series is sponsored by the Graduate Center’s Office of Career Planning & Professional Development.
MISTY CROOKS, HOST: I’m Misty Crooks, a PhD candidate in Anthropology at the Graduate Center and a fellow in the Office of Career Planning and Professional Development. In this episode of Alumni Aloud, I interview Chris Arettines, who graduated from our program in Mathematics and is now a Quantitative Developer at BP. He talks to us about translating your graduate school research skills into industry success and how to develop a successful career in the world of data.
Chris, thanks for joining us today. I’d like to start out with a couple of general questions. You currently work for BP as a quantitative developer. Can you speak a bit about your organization and how your role fits in there?
CHRIS ARETTINES, GUEST: Sure. BP, as you probably know, is an energy company. They have a lot of different business units. The particular business unit that I’m in is their trading organization. Their trading organization, you should kind of think of it like a hedge fund or some other kind of trading firm that takes speculative risk to hopefully try and make money. And we are a pretty large component of the broader BP organization, but my group is specifically about trading. That’s what my unit is. I am a quantitative developer, and what that means is I’m the guy who’s helping to do research and coding up quantitative base trading strategy, so that means we use various statistical tools and algorithms to decide how, what, and when we trade and then we need to understand if this is a good idea or bad idea, so we need to build lots of tools to analyze those various approaches. In a nutshell that’s kind of what I’m doing inside of the broader organization.
CROOKS: Okay, great and you’ve actually held a variety of types of roles in your career. You’ve been a data scientist, a research engineer, and now your current role. Can you tell us a bit about your journey from graduate school to getting your current position?
ARETTINES: Sure, I was in the math program at the Graduate Center. As I was kind of approaching the end of writing up my thesis and getting ready to defend, I was looking at the job market. And it’s very difficult to get the kind of postdocs that I wanted to get at the end of my kind of academic journey. I would have to move to other states or other countries to get a postdoc and I was in a serious relationship at the time with the person who’s now my wife and she was still in school. I didn’t really want to like just have to go across the country to obtain a position, so I started thinking, Okay, what are some jobs I can get in New York City or near New York City with my kind of background. And data scientist was something I’d read a lot about in blogs, and you know just generally heard about lots of other mathematicians, statisticians doing, so I looked into it. I you know brushed up on my programming skills. I’ve done some programming as part of my graduate research as well, so you know I wasn’t starting from scratch. But I brushed up on that. I brushed up on statistics which wasn’t really part of my PhD. My PhD was in hyperbolic geometry, very little statistics involved there, so I brushed up on all that, you know machine learning, all these buzz word type fields that I had been reading about online. And then I went to various job application portals and I just started submitting my resume while reading up on interview questions and practice problems and things like that.
And so, I had taken a teaching job in the evenings just that I can make some money, but with the expectation that I would eventually get a job as a data scientist during the day. Eventually I interviewed with the right place and I got a job offer as a data scientist. And for those who don’t know, a data scientist really it could be, can mean many different things. But in my case, it was somebody who was doing statistical research for an ad tech company that was trying to figure out how to optimize the ad content on behalf of my company’s clients. There’s a lot of mathematics involved. There’s a lot of modeling involved for those kind of roles. They look for people with quantitative backgrounds like math PhDs or physics PhDs etc. So that’s how I got into the into the field.
CROOKS: I think you’re hitting on some important things there: the steps that you were taking to make yourself a viable candidate for these roles and also the personal component that we deal with as students when we’re trying to make these choices. I think that’s really helpful. In your current role, what’s a typical day or week like?
ARETTINES: I work on big infrastructure projects that my trading team uses to kind of analyze the trading ideas that they have. Say some other researcher comes up with an idea, hey if I buy and sell this contract, according to this algorithm, how does that look like going back ten years. I basically built you know infrastructure so they can kind of plug that idea into my system, look at the impact of how profitable or terrible that idea is, various you know statistics related to the performance of that idea. And so I built the platform that can do that, so some of my work involves just like adding little bits and things here or there to that. Sometimes I actually do the analysis of an idea myself like I come up with hey, what if we did things this way instead of this way. Let me code up something and see how it looks and then if it looks promising, we’ll take it to the next level and we’ll do a deeper dive and we’ll talk with the traders and all this stuff. So it’s really a mix of programming, big system work, and having some research ideas, following through with them to see you know if they work or they don’t work.
CROOKS: It sounds like there’s a good amount of team-oriented work. Would you say that overall there’s a lot of that? Is there also some autonomous side of the job as well?
ARETTINES: One of the things I like about my current job is there’s a lot of both and I get to kind of choose kind of the balance that I want on any given week. For instance, if I’m developing this infrastructure, I designed it. I built it from scratch. I can kind of add things to it if I want to, and I have the flexibility to tell my team, hey guys leave me alone for a week. I gotta work on this thing and they’ll leave me alone. But on the other hand, there’s also a lot of opportunity to collaborate. We have weekly meetings with researchers, traders who all have very different backgrounds, and so we have very vibrant discussions about what we should be doing next or what we should be looking at. So really there’s a good mix of both of those modes of working, which I appreciate.
CROOKS: You touched on this a little bit. What do you enjoy about your job? What do you find the most rewarding?
ARETTINES: I really enjoy being able to design and build the system that I built from scratch and own it myself. In my team at my current company, I was responsible for kind of building a lot of that stuff myself. It’s kind of fun to be able to design all that infrastructure from the ground up and own it instead of having it handed to you and you inherit it and you don’t necessarily understand all of it. When you build it yourself, you understand it, and I like that aspect of ownership that I have in my current job.
CROOKS: What are some challenges that your organization or even perhaps your industry as a whole is currently facing?
ARETTINES: I would say that the thing about the finance industry and the trading industry is that it’s a dynamical system that’s constantly evolving. The conditions in the market are changing. The people in the market are changing, so it’s never going to reach a state where it just kind of like stabilizes and becomes fixed. It’s always going to be in flux because there’s human component and the global economy plays into every aspect of the markets, so it’s never going to be a dull time when you’re working with financial markets, so that’s very interesting as well. So it provides challenge but also opportunity because things are constantly changing. If you don’t change, someone else is going to come in and, and you know eat your lunch. So it’s an interesting field that’s always changing and very dynamic.
CROOKS: And it sounds like, then, that pushes people in that industry to be constantly learning as well.
ARETTINES: Yes, and that’s another thing that I really enjoy about not just this line of work. As a data scientist, it was the same thing. You always need to keep up with the latest trends in research, the latest advances in machine learning or cloud computing to figure out how are the ways I can apply those ideas to the problems I encounter at work. So yeah I’m always learning, whether it be about markets or technology, or you know just kind of more abstract things, like about system design. I enjoy that aspect of always learning.
CROOKS: If you were going to give some advice to people looking to looking to go into that field, what would you say are the keys to success?
ARETTINES: I would say, for anybody say, coming from a quantitative background, to enter into the space of data science or financial research, you need to know how to program. That would be the single most important advice I would give to anybody thinking about the field because pretty much, there’s nobody these days who’s a researcher who doesn’t also know how to code, at least basic stuff. You should pick a simple language like Python or maybe R, something like that. Learn the basics. Learn how to ask questions. Learn how to answer questions using those languages because in any job that you take on these days, you’re going to be coding up somewhere. So that’s a very important skill to develop going into the job market.
CROOKS : And you talked a bit about what your job search looked like. I wonder if you can give students advice about specifically interviewing, how you translate yourself for employers?
ARETTINES: One of the things that I did to help motivate me and also to just learn new things was I would come up with like projects that I thought were genuinely interesting that would force me to learn how to do something I didn’t know how to do. Like I figured out how to scrape data from the MTA’s website and make these visualizations that I put on my website. I learned a ton doing it, like looking back now, and now that I have some professional experience like oh that was so stupid, why did I do it that way. But like just the exercise of trying something new and having an artifact that you could show an employer, as a result of that, was a very valuable experience and I tried to repeat that several times as I was kind of building out my resume and building out my experience. And I think having these projects and like a portfolio to point to can be very helpful for pitching yourself to an employer. Say hey I didn’t know anything about this, but I taught myself, and I was able to do xyz. I’ll do the same for you. Even though I might not come in knowing everything there is to know about online advertising or the natural gas futures markets or whatever. If you can demonstrate to these people that you’re a good learner and you can build things with your knowledge, then that’s a great way to pitch yourself.
CROOKS: That’s really interesting. Related to that, is there any advice you could give current students about networking?
ARETTINES: Yes, so another thing I would recommend is: When I was at the Graduate Center, people from different industries, I remember two that I went to that were people from the finance world, just came in and gave like an hour-long talk to whoever was interested about what it’s like in the field, what are the skills that they look for. And I tried to go to a few of those. I would also look on, I think, maybe it was meetup.com or some other website where you could see, oh there’s some free networking event happening at Bloomberg. The way these events usually work is there’s some period of time of just socialization. Then you listen to a bunch of presentations. Then there’s another hour of socialization and I would just you know, I’d have my little cheap business card that I printed out with my email on it, and I would just talk to people and if they seemed like they were working on something interesting that might be aligned with what I was looking to do, I’d give them my business card. I don’t think I ended up getting anything from that, but that’s the kind of mindset that is good to have I think to get your name out there. Because, with a lot of these kind of online like job submission portals, you’re a few kilobytes of data. There’s nothing that necessarily makes you stand out. But if you actually hand your business card to somebody who knows somebody in the recruiting department, then that’s like a way you can potentially get in the door. And getting in the door to an interview at any of these places is half the battle at least maybe more. That’s definitely something I recommend is going to these networking events and having some way of giving them like a business card or have a website you can point them to with your contact information on it or something, something like that. Lots of different companies host various meetup events all the time, find one that’s interesting to you. If you can register to it, just go there, meet people. It might lead to an interview opportunity, it might not but, at the very least you’ll increase your network size, which is a good thing.
CROOKS: What are the traits or qualities you think that are useful for somebody going into your field?
ARETTINES: Definitely a quantitative mindset. The ability to communicate is also extremely important. I’ve worked with people who are extremely brilliant. But if they can’t communicate, it’s just hard to work with that person. So ability to communicate, which is something that we learn as GC students because of all the teaching we do is, I think, very important. Quantitative mindset. And if you can solve problems independently, without having your hand held the whole time, I think that’s also good, and I think that’s also one of the things you learn as part of a you know say PhD program where you need to kind of take ownership of your own research and see it through. So I think those characteristics serve you well in this kind of field.
CROOKS: What do you mean by quantitative mindset?
ARETTINES: I’ll give you an example from like kind of my typical workday. A trader comes to me. He is not a statistician. He’s not a computer scientist, but he knows how the markets work, and he has an idea to explain to you. Is there some way we could turn xyz into a profitable trading strategy? Now you as a quantitative researcher, quantitative developer, need to understand: How do I translate this problem expressed to me at a very high level from a human who doesn’t know about four loops in Python or whatever? How do I translate that problem expressed to me into the language of math and a computer, so that I can code it up and then evaluate it very rigorously? And being able to do that translation from business level problem to mathematical or computational implementation is kind of what I mean, being able to do that translation seamlessly. At the same time, you also have to be able to communicate what a bit of code is doing to a trader because they’re not going to accept your word for it. Oh yeah, here. Here’s a black box. It’s going to make you a million dollars, I promise. They’re just going to say no, like you have to explain what it’s doing, and you have to be able to explain it and answer questions about it, and these other things so communication, both understanding and communicating to others, is also very important.
CROOKS: And it sounds like also being able to get people on board with your ideas and what you’re doing.
ARETTINES: Yeah, yeah like that’s, that’s something that you learn in the business community that you don’t learn in grad school is like you understand that there’s no escaping some amount of politics and like maneuvering that you have to do to get people to agree with your point of view. When you’re a researcher, you’re not necessarily too worried about those things. You have a problem, you have a theorem to prove or you have some problem solve. Enough people are interested in it. You go prove it. In the business world like you have an idea that you want to take into your product or turn into a trading idea, and you have to sell it and you have to get all the different people who are stakeholders to buy into it. And so you have to be a bit of a salesperson sometimes. And so yeah that’s definitely also part of just any kind of professional environment. You have to know how to sell your ideas to the right people.
CROOKS: In thinking about the future of the field of data science, quantitative development roles, what do you think that looks like? Where do you see that field going?
ARETTINES: I think the demand for that kind of person will only grow. The world is becoming increasingly automated. The availability of data is constantly growing exponentially. The amount of data that was publicly scrapable 20 years ago compared to today is astronomically different. And so the world needs people who can understand this data, ingest it, understand it, and produce meaningful insights. And this applies across every single industry. That has nothing to do with finance or ad tech. Every industry has large volumes of data that they’re trying to understand, so people who have the kind of technical knowledge and ability to manipulate that data and understand it, and then communicate that understanding, are going to be more and more valuable.
CROOKS: Thinking about how current students can work towards this career path, are there professional skills you would recommend people develop?
ARETTINES: Definitely start learning basic programming language, if you haven’t already, if you don’t know one. I recommend Python just because it has a relatively low barrier to entry and generally if you can code in Python, companies won’t … like you don’t need to know C++, a very low level language to get a job at the kinds of places that I’ve been talking about. If you know Python and can do a decent amount of data analysis with Python, that will get you pretty far in terms of the minimum skill set required to be able to get an interview and start to you know make inroads with these companies. And now, in order to do that, if you don’t have the experience already, I know some people have done boot camps. There are these boot camp programs where say after you finish your program, you’ll go to one of these boot camps for say two, three months or however long it may be. And it’ll be like military boot camp where like you start with nothing and you end up being reasonably competent at these general skills that companies are looking for. And these boot camps will help you, try to help you find a job at one of these companies because they basically are incentivized by the relationship with a potential employer to provide them with good employees. So they make money, you get a job. I know some people who have had success with these programs. Others have not, but you know, like it’s definitely something to be aware of.
CROOKS: Relatedly, we’ve talked some here about sort of how you supplemented what you were doing in grad school to get to where you are. What would you say were the skills and knowledge you gained through your experience of being a grad student that helped you?
ARETTINES: One is definitely communicating, so I taught a lot. I gave talks at various seminars just to share my research, but you know you also learn, yeah this works, this doesn’t work. I should explain it this way instead of that way. You accumulate these lessons the more public speaking you do. I think that was very valuable to me. And just being able to have some self-direction to my research. You know I would obviously meet with my advisor regularly and we would have back and forths about you know what was working, what wasn’t working. But a lot of the time, at least in mathematics, you’re just kind of by yourself, trying out ideas and a lot of the ideas don’t work. You thought you had a way to prove the theorem and turned out it didn’t work because of xyz. You just wasted a week, so you have to go back to the drawing board. There’s a lot of independent problem solving as part of a PhD program that is, I think, also helps build your perseverance for jobs outside of academia as well. I think that’s why companies like to hire PhD students because nobody hired me thinking that what I knew about hyperbolic geometry was going to be useful. The PhD designation is more to say like okay this person knows how to do research. They know how to independently problem solve and they have the perseverance to see the program through to the end. So I think those are really the characteristics that that you can bring to the table as a PhD student that somebody with just say undergrad experience doesn’t necessarily have.
CROOKS: That’s really helpful and I think that’s a great way to frame it for current students and have them think about what they’re bringing to the table. To wrap up our discussion, I’d like to ask if you could give any piece of final advice for graduate students.
ARETTINES: The role of academia is in one sense very relaxed and the other sense very cutthroat because there’s so few positions. The reality is there are too many PhD students compared to the number of positions in academia for PhD students. It’s inevitable that people are going to have to potentially face the reality of giving up their dream of being a professor or something like that and going into business. You should ask yourself honestly, are you okay with that. For me, it was very troubling when I was in the last step of that process and I realized I’m not going to be a professor after all. If I had come to terms with that sooner, it might have been a little bit easier of a transition for me. You know, I love what I do now. I’ve had a very fulfilling career so far. But you know, like at the time it was a little bit jarring to be, to confront the reality of yeah, I’m gonna have to leave academia. There’s just nothing I can do really. I’d have to sacrifice too much to stay in it. You know, that won’t be the case for everybody. Some people will be able to get positions no problem, but the numbers don’t lie that there’s not room in academia for all the PhD students. So the sooner you come to terms with that reality and make peace with it, the easier psychologically your transition will be and your planning will be to you know explore alternatives deeply sooner.
CROOKS: That’s really, really helpful. And this whole conversation has been really helpful and really enlightening and I thank you for joining us and giving us your wisdom and expertise today, Chris.
ARETTINES: Yeah, of course, thank you for having me. I’m always happy to share.
This entry is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International license.