The Beginning

Aug 06, 2023

Quantiative finance field is full of eager people. People that expect getting 6 figures in their anual checks. There is no anything bad with this, we al work for gaining our deserved payment. And even though I had this way of thinking when youger, I’ve also learned to enjoy the journey of learning about models, techniques, and challenges that push you beyond your “limits” from time to time.

Quantitative finance comes as a way to move from pure software development. I used to work as a software developer, and I liked. Nonetheless, as software engineer I experimented the monotonous pain of doing similar things many times over the week. I browsed the job market around 2011-2015, and the topic didn’t change to much. Somedays solving bugs, another ones addig features to the system but not always the most interesting ones. In my experience, the best one was being the support of people with better knowledge than me by leading projects from technical perspective. Unfortunatelly, still having this really nice experience I felt incomplete. I have the impression that it was also because at that moment I has known about quantitative finance.

I ought to confess that first semester as a systems engineer student I was tempted to left the program. Why?, well, the naive idea about C++ programming language useless, if Java already exists. Now I’m grateful with people that motivated to stay there, both direct and indirectly. Having left the program would represent my profound regret nowadays. As many of you know, last 20 years C++ has been the core programming language for pricing derivatives. It has the power that no any other programming language has, even Java. I’m aware that it is ugly if compared with others specially on those days, but still powerful. Now with C++ 23 this perception may change. Anyway, the experience during the bachelor degree in system engineering has contributed a lot to my current development as a quant. Other programming languages have raised, as python, rust, Julia, Ocaml, Haskell, among others, but C++ is still the king for real computation. However, don’t get me wrong, I also know python and I like, just that has complementary application. Python has its own merit in my career development since it was plus former bosses pointed out when I joined first time to pure finance corporations. Many people knew the fundamentals, but I knew how to produce industrial software, totally different. And with this I just want to highlight the opportunity that this knowledge brings to quants. Knowing how to make your code good enough to be in a production environment is something that you gain only with experience, not with books or bootcamps. I now have expectections with LLMs and RAGs, since they have the potential to boost this evolution. I hope not soon that skill may be comparable to Fortran at 2000s.

Do not forget, ever the quantitative stuff, that’s why people refer to us as quants. A quantitative analyst must know the fundamentals in probability, statistics, finance, mathematics be cause that is what we do during the day, assessing and modelling systems with numbers. People that knew me in university (systems engineering) may aknowledge that I was partially interested, because at that moment even physics seemed to me a very far approximation of reality. I think the root cause of that was my experience learning about the quantitative approach of electric fields in my technical highschool. But a miracle happened!. A friend motivated me to participate in a trading concourse performed in Colombia, anually. If you have already read the post about me, you may infer there were many events happening around 2011-2014. Well, this is the second driver to get involved in finance, the deep interest for knowing how market agents do shape prices. Of course, in that moment I didn’t think in so technical words, I was just interested on what’s the reason the prices move so fast in one trading session. As result, I started my second bachelor degree in economics, from scratch, I didn’t want to learn everything as any other student, even though I was almost finishing my bachelor degree in systems engineering. As a result of this overlaped experience between this fields of knowledge I started a pretty simple model that afterwards turned into my undergradate document in systems engineering, titled “A FX strategy using graphs theory”. This model tried to to solve the problem of getting the best route to take advantage of arbitrage in the market given crossed quotes in the market. Currencies are the nodes, and quotes are depicted by the edges, the rest is your imagination. Unfortunatelly, in that moment I coded it in PL/SQL, and please don’t ask why, I just did it. Writting this lines just made me reconsider rewrite it again, with market friction parameters to see what happens. Nice material for a new post.

So, wrapping up, the convergence was not planned, as it was not the original idea to become a quant before second time joining university to obtain a diploma as a economist. But I think I have been blessed, or lucky as you wish name this, because many thinks have happened and every day I feel more into my profession as a quant. I do not forget the discipline that requires to know as much as you can of different branches of knowledge to be creative and innovative. Nonetheless, I’ve learned that there is a point in which is clever to start specializing. I think I do have a clear perspective of what I want to be specialized in. What do you guess it is?. A hint … it is not to be a software developer again, I’m greateful because of the experience in the field though …