How did you become a quant?

Question

_{This question will serve as the definitive community wiki of career anecdotes. Future career questions can be pointed here.}

There are as many career paths as there are people. How did you get started in the industry?

Answer 1

I first found-out about quantitative finance during my PhD program. Like most grad students, I was looking for an excuse to procrastinate from my dissertation. I began looking at the course catalog of different departments until I saw a course entitled "The Mathematics of Finance". The syllabus listed terminology I had never seen before, like Black Scholes. When I read-up on what options pricing was, I realized that this was an excellent application of my academic specialty: high-performance computing.

I applied to every job ad at an investment bank or hedge fund that listed a PhD in computer science. I eventually got a position doing statistical arbitrage on a proprietary trading desk at a bank. So began a ten-year journey of stat arb and HFT at various banks, hedge funds, and small prop shops.

Quantitative finance is a career of constant learning; I have had to continuously absorb new technical skills and domain knowledge. It's also a field with potentially unstable employers. The first three trading desks I worked at all shut down, which is why I have an enormous amount of skepticism for most firms in this industry.

Finance is ultimately a game of infrastructure, and most places (and people) are surprisingly bad at this. The only competitive advantage is time. As in, how much time did you spend practicing your craft? How much time did you spend building parameter optimizers, data pipelines, etc.? That's why I have no patience for people who claim they have to keep everything secret; almost no trading idea is transferable from one firm to the next without an enormous investment in infrastructure.

So for anybody who wants to enter the field, I would suggest that you learn as much background material as you can, including general finance, accounting, economics, etc. Also, get really good at programming. (The language doesn't matter; your boss will tell you what to use.) And while you will have to practice your math, don't get attached to any single technique or asset class. Ironically, despite learning of this industry because of Black Scholes, I have never traded options professionally.

Answer 2

personally studied physics and ultimately converted to finance like many simply looking for a job.

just to be rigorous as this is what it's all about in this field, quant is not a job, it is more like a label.

it can mean pricing derivatives for the front desk of an investment bank, validating macroeconomic stress testing models in its risk division, doing bayesian inference for a statistical arbitrage fund, or even just developing HFT microstructure arbitrage algorithms, that have absolutely nothing to do with finance whatsoever.

financial modelling could mean looking at financial statements in Excel and doing some basic accounting maths in an area like M&A or credit research, or trying to fit a complex stochastic quantitative model to a particular pricing or volatility clustering problem.

the first step is understanding how it all ties together, who does what, banks, exchanges, brokers, funds, regulators, central banks, and the professions within them traders, market makers, structurers, strategists, economists, etc. it can take years..

then decide which career is right for you, acquire the specific skills if you don't have them, and be quick to land a job, to learn and adapt, the market is a constantly evolving place of survival.

Answer 3

The late Mark Joshi wrote this rather famous advice on 'How to become a Quant'.

Answer 4

I am coming from a "soviet rocket science" family, my dad was doing military aerodynamics (soviet Space Shuttle, hypersonic flows etc) and so I thought I would.

With Russian economic reforms in the 1990s it became clear that to maintain the social status in the society (which was very high for the rocket scientists) you had to go into commerce, ideally banking. So I dropped science without a second of regret, switching to CFA-sort of finance, ending up doing MBA at INSEAD after then 1998 default.

I didn't take a single course on derivatives there, thinking that using exponentials in finance is an utter time wasting rubbish, until somebody told me that there were physics PhDs in banks who made lots of money and directed me to one of the early Wilmott's books.

Heard the calling, did a U-turn and the rest is history. It was much easier to get an entry level quant job 25 years ago. I did a part time PhD in stats later, part time, for pure fun. It was never a requirement to have it, except for few teams in few banks.

To anybody who aspires to become a quant, I would suggest to be honest with themselves about what they are trying to achieve and why. Academic programs don't actually reveal what you will be doing day on day.

Modern quant jobs are mostly (software) engineering. There are effectively no theoretical jobs, and pay correlates with your ability to program, so C++17+, prod-grade, or Python at least. PDEs are somewhat important, MC is very important. There is nearly zero opportunities to write theoretical academic papers; there are opportunities to do less coding and more paper work, but mind the earlier comment on pay. Get ready to never use Ito Lemma: with high probability it has already been used in the problem, solution to which you will be supporting and extending decade(s) ago.

ML etc is used very little in the Markets operation, but it does gradually crawls in via electronic trading, autohedging etc. There is more demand and, actually, less demanding ML jobs on the buy side. But of course, comp volatility is much higher, therefore it is not obvious that short term Sharp of your comp will be higher that on the sell side, unless you become at least a sub-PM, or a partner some other way.

The single most important thing, that separates successful banks/qa teams and not so successful ones is taking the applied part of your job seriously and having general problem solving attitude. You see a "challenge" - you own and fix it. Teams that don't nurture this attitude ultimately fail. So don't hope for an easy ride.