I was emailed recently with a career-related question about jumping from one quant role to another. The question posed was "How can I make the jump from being a quant/software developer to a quant trader/researcher in a fund or investment bank?". This is certainly possible and does happen occasionally. However, it will require some extra-curricular work, and some initiative, in order to make it happen.
The main issue is that the educational background, as well as professional experience of quant traders/researchers and quant developers is quite different. Hence if you wish to successfully make the transition you will need to gain additional skills that you might not otherwise have picked up.
In addition, you will need to convince your team lead or another team lead, that you are sufficiently capable of carrying out independent research into quantitative trading strategies that can produce "alpha" for the firm you are working at (or wish to apply at).
Assumptions About Background
As always with the QuantStart Mailbag questions I need to make a few assumptions.
Firstly, I am going to assume that you have a "traditional" software developer background and are likely to be formally educated in a university setting, at least to undergraduate or masters level.
Secondly, I will assume that this will be in a Computer Science or Electrical Engineering related degree, although also in Mathematics, Physics or alternative Engineering discipline.
Thirdly, I'll assume that you do not have a research level qualification, such as a Doctorate (PhD) or Masters in Research (MRes). If you do, then the transition is likely to be far easier.
Hence you are likely to have a strong knowledge about technical architecture, software development, best practices (version control, testing etc), database design, interaction with service APIs, communications protoocols and proficiency in a language such as C, C++, C#, Java, Scala, Python or even a functional language.
It is less likely that you will be well versed in probability theory, statistical methods (Frequentist and Bayesian), econometrics, derivatives pricing or probabilistic machine learning.
However, you will certainly have a good understanding of financial markets if you are building quantitative trading infrastructure. You'll know about sending orders and receiving fills, and may have some exposure to actual trading strategy code implementation, perhaps as you've been asked to optimise a prototype algorithm.
Learning The Necessary Skills
There are three paths that are most likely to lead to working as a quant trader if you've already been a quant developer. The first is the research route, the second is the extra-curricular route and the third is via prop trading.
A good option for ultimately becoming a quant trader is to head to grad school and study an important field for quant trading research such as:
- Probabilistic Machine Learning (PML)
- Deep Learning (DL)
- Bayesian Statistics
You could either work directly in a Computer Science or Statistics department (or interdisciplinary departmet) for a more theoretical treatment or perhaps head to a Mathematical Finance department and work directly on applying PML, DL or Bayesian Statistics methods directly to financial markets. The latter is attractive, particularly in the top schools, because your department will likely have strong links with funds and IBs that want to make use of your work.
However, this will mean leaving industry and returning to academia for three years and more likely four, at least in the UK. In the US this could be up to six years. Fortunately, the study of such techniques is highly engaging and will likely be very enjoyable, as well as intellectually demanding. But you will suffer the consequent lack of earnings, especially if you are used to a comfortable fund or IB developer salary.
An alternative path in to quant trading is to simply make the transition internally at your current firm. You could also apply for a quant trading job directly at another firm, bypassing the PhD/Masters route.
However, this will likely be harder as you will be competing with those who already have research degrees and an academic publication record. Hence you will need to demonstrate ready made quant trading skills that can be directly applied to the firm immediately.
The best way to go about learning these skills is to dive into retail quant trading at home. Initially there is no need to commit any capital as you can learn a lot of the techniques by "paper trading". That is, calculate the profit and loss of your strategies by seeing what you would have made if the trades had actually been placed. This will give you a good understanding of what the mechanics are like.
QuantStart specialises in quant trading for retail quants and there are many resources on the site discussing just that. The best place to start is the Beginner's Guide To Quantitative Trading.
Since you are already a quant dev perhaps the best way to learn is to try understanding one of the many backtesting frameworks available such as QSTrader, Zipline, PyAlgoTrade or PySystemTrade. If you already have knowledge of such systems through your day job as a quant dev, then you can dive straight in to strategy research using one or more of these tools.
You will also likely need to understand the basics of Time Series Analysis, Bayesian statistics, Machine Learning and Deep Learning if you wish to utilise the latest techniques for quant trading strategies. This sort of self-study can be carried out at home, or during lunch hours of your day job.
If you feel you don't have the necessary mathematics background for these topics, then you can also begin learning it yourself in your own time via the following articles: How to Learn Advanced Mathematics Without Heading to University - Part 1 and How to Learn Advanced Mathematics Without Heading to University - Part 2.
The main skill that quant researchers possess is the ability to come up with new trading ideas based on perceived or identified pricing anomalies within certain markets. Having the ability to mathematically model an entirely new situation is precisely the skillset that is obtained from carrying out a research degree. If you can learn this ability on your own, then you will be competitive.
The advantage here is that you'll already have professional level programming skills, which is often a stumbling block for those coming from an academic environment where software engineering capability is often not as widely present.
A third alternative is to join a proprietary trading ("prop trading") firm. However, you will likely face a similar scrutiny to a hedge fund interview since the firm will be committing capital to your strategy. In addition it is rare for prop trading firms to front all of the capital these days, so you will likely need to have a prior track record with your own capital before applying.
Applying For A Role
Once you believe you are sufficiently capable of researching new strategies it is time to apply for some roles or even ask your team lead if you can make the transition.
There is likely to be resistance initially to the latter, but if you can demonstrate alpha generation with some solid backtests, or even an actual track record, then you will be in with a strong chance.
Ultimately if you feel your current firm is undervaluing your skills then you should always consider applying externally. Your extra-curricular CV will almost certainly help you stand out in a crowded marketplace.
If you have any further questions about transitioning to a new career role, or otherwise, please feel free to email the team at email@example.com and we will get back to you as soon as possible.