I finished my academic career back in 2017 with a degree in Physics moved to Chicago, the center of derivative trading, and became a quantitative researcher with a rapidly growing trading firm. Since then, I have been frequently asked by friends, alumni and even random applicants on LinkedIn on how I did the transition. Sometimes, I would even get this question from candidates at the end of an interview (me being on the easier end, of course).
Contrary to the common misconception of all quants being math geniuses who solve Rubik Cubes when they are drunk, it is actually much easier to pass a quant interview than a lot of people might imagine if you spend your time on preparing for the right questions. I will share some of the things I have done below.
No Financial Knowledge Needed
Even though quants are making and losing money in the financial market, a deep understanding in finance is not necessary. When I was interviewing with proprietary trading firms and hedge funds, I had close to zero exposure in finance. I knew what a futures contract was, calls and puts are the most common types of options and that’s pretty much it. As ignorant as I was, I didn’t even know how Finance differs from Economy.
Highly Advanced Math Not A Must
While there are many well known giants in the field hold advanced math degrees (e.g. Jim Simons from Renaissance Technology), there are in fact more physicists and computer scientists than mathematicians in this industry. From my personal interviewing experience, candidates with a solid understanding and grasp of undergraduate level math are usually more than qualified to solve the interview questions. Just think about it this way, how can anyone be expected to solve Goldbach’s Conjecture within a one hour interview?
Great Coders Celebrated
It is, however, crucial that the candidate is a good coder. Quant researchers communicate, and realize their ideas using code. Any analysis I perform, any model I derive, and any strategy I design are done using one or more programming language. As a quant, I literally spend more than 50% of my time in the job writing code, another 20% reviewing other co-workers’ code. Being able to turn a good idea into production code that is robust and bug-free, and also easy to build upon, is a key requirement of a good quant researcher. Latency sensitive and performance oriented prod/backtest code is usually written in C++, while data analysis and visualization tasks are carried out using Python.
Most Crucial Skills: Problem Solving and Continuous Learning
While knowledge of specific techniques or approaches are sometimes important, these are far from being the top skill of a quant researcher. I have been asked more than once in my interviews if understanding of a specific machine learning algorithm is necessary (thanks to the booming of ML in the past few years!), or if we need people who possess in-depth knowledge of a certain optimization technique. My firm belief and my answer has always been that all techniques/algorithms are simply part of the tool box and of the researcher’s discretion and disposal. Knowing what is the best technique to solve the particular problem in hand, and acquiring the necessary knowledge in the process to apply that tool successfully, is the core competency and separation of a great quant researcher from a good one.
This article aims to be the first of a series I am hoping to write to uncover careers in quantitative finance, and please feel free to ask me any question you might have!
