Oscar Ma
Derivative Trader at Valkyrie Trading
What inspired your interest in the quant/data science space, and how did you end up in your current role?
I got hooked on quant pretty early. It felt like the grown up version of chess and poker, just with more screens and slightly higher stakes than bragging rights with friends. At Swarthmore I fell into math, stats, and computer science, and MSCF was where that turned into a more structured obsession with markets, algorithms, and data. Through a trading bootcamp at Valkyrie and a quant research internship at PIMCO, I saw both the research and trading sides of markets. That mix is what ultimately brought me back to Valkyrie, where I now market make options on the Treasuries desk.
How did the MSCF program prepare you for the technical and analytical demands of your role?
MSCF definitely gave me the technical toolkit: data science, stochastic calculus, derivatives, fixed income, optimization, all the ‘buzzword’ you actually need to survive on a desk. But more importantly, it taught me how to learn fast without drowning: how to pick up a new model, a new product, or a new dataset and get from “no idea” to “useful” efficiently. Constant Python projects, late night group work, and trading competitions felt a lot like the job now, where you rarely have perfect information, but you learn to prioritise, structure problems, and ship something that works under time pressure.
Looking back, which MSCF courses or professors had a lasting impact on your career?
Two big influences were Professor Chad Schafer and Professor Ron Yurko. Chad’s Probability Prep and Financial Data Science I & II really shaped how I think about noisy, real-world data, which is basically all market data. Ron’s Machine Learning and NLP courses pushed me to build models that actually hold up out of sample and to explain them clearly, both of which translate directly to trading and research on the desk.
How do you stay current with evolving technologies and research in the quant space?
I mostly stay up to date by staying hands on: building backtests, dashboards, and tools on real trading problems forces me to learn new libraries, methods, and tricks quickly. Around that, I try to keep a light but steady diet of industry blogs, conference talks, and plus the occasional “someone on the desk dropped a link in chat and now we’re all arguing about it” thread. Honestly, the best filter is still talking to other traders and quants, if an idea keeps showing up in discussions and in PnL, it’s usually worth digging into properly.
What advice would you give to students who want to pursue a hybrid quant/data science career in finance?
I’d say you should aim to be strong in three areas, coding, statistics, and markets. Build real projects such as backtests, small trading bots, and dashboards, instead of just polishing your resume. Try to intern in different environments if you can, (prop firms, asset managers, or fintech), so you can see what kind of pace and culture you actually like. Finally, do not underestimate communication. Being the person who can translate between traders, quants, and devs is a huge edge.
What’s something you've learned on the job that challenged your expectations or assumptions from grad school?
One big thing I learned is how important team based trading actually is. In school you mostly think about your own PnL or your own model, but on a desk a lot of the edge comes from communication. Talking through views with other traders and making sure we are not all leaning into the same risk often matters more than any single clever idea.
The second lesson is how real tail risk feels when you are actually holding positions. During my solo crude oil rotation around Trump’s liberation day, the market put in a move that was close to a four standard deviation event. On paper that type of move should be well below 0.1%, yet it unfolded in real time on my screen. That experience heightened my intuition from grad school and reminded me that managing risk and stress testing scenarios is just as important as everything else we do on the desk.