An interview with Renos Karamanis, Simudyne intern

How did you find out about Simudyne’s Internship programme?

A couple of years ago I started looking toward quant research positions for my next professional destination after my PhD degree. I was in the midst of the job application process when a very good friend of mine recommended Simudyne as a more attractive alternative to working for a major investment bank. Applying for a quantitative modelling role at Simudyne immediately seemed appealing to me due to my agent-based modelling background and the potential opportunities for more diverse technological applications.

What projects have you been working on?

So far, I had the chance to be involved in a variety of projects, stretching from quantitative finance, such as meta-order execution strategies in trading, to public health projects such as modelling the spread of Covid-19. Every project has an agent-based modelling component which includes agents with calibrated behaviours usually conforming to non-trivial mathematical models. What is compelling is that we get to shape these models based on detailed discussions with clients and comprehensive literature reviews to understand promising areas of application.

What were your experiences joining the team?

Before joining, I was looking for an opportunity to learn and become a better version of myself as a researcher. I believe that Simudyne turned out to be that opportunity. Over the first few weeks I had to read a lot, but I found that my colleagues were always keen and available when needed for support. Most importantly, I find myself surrounded by highly competent professionals who can prove to be a valuable resource if I ask the right questions. I also perceive the culture of the company to be very appealing, as although people are highly competent, they are not competitive with one another, which in my opinion contributes to a sustainable working environment.

What do you feel you will take away from your internship?

I think firstly, it will be the ability to instantaneously switch between contexts. The ability to read something and put it in place quickly is an important skill. In essence, this can be summarised with one word: adaptability. I will carry this into the rest of my career.

Another thing I regard as valuable experience is the level of precision and realism that is intended when structuring a model which emulates a real system. This also includes the amount of scientific scrutiny that follows its implementation. I feel that these modelling practises are extremely useful for any type of quantitative role across many industries.

What do you enjoy most about working at Simudyne?

That would be definitely how exciting the projects are. You get the feeling you are having a major impact on the client and ultimately society. A contributing factor to that feeling is that the stakes are very high. To put this into perspective, depending on the amount of effort and ingenuity you put into the work, you could be a part of a major movement. By movement I allude to a widespread use of agent-based modelling throughout many industries. In summary, I enjoy the idea of being part of this movement and this is mainly possible because of the inclusive culture of the company.

What did you find most challenging?

The context switching between clients and industries and then capturing it in a model that is representative of reality can be hard. You want to build a mathematically state-of-the-art model and reconciling that with reality is an achievable, but difficult intellectual challenge. Accelerated literature reviews can also be a challenge. In contrast with a PhD degree where you dive into the specifics of a problem for an extended period of time, at Simudyne, you have to identify, understand and implement the state-of-the-art at a much faster pace. Nonetheless, a helpful factor to this process is the insightful discussions we have with clients, and the collective experience between quantitative modellers across different fields.

How did you overcome any pandemic related unintended consequences?

I have to confess, initially I thought it would be weird joining the company from home. How would you show a newcomer the existing systems? How would you ask a question if you are not in the same room and how would you learn from others from home? Despite of these anticipated challenges, the induction was quite smooth as the technological solutions for remote working were already seamlessly embedded into the working routine of the company. Don’t get me wrong, not having to commute, but instead waking up and going to my laptop is extremely easy, but I do miss the variety of the day.

Beyond the uses clients already have for agent-based simulation, how would you apply it to a problem?

I have been thinking about applying it to capture the dynamics of, and potentially resolve international conflicts. I believe it would be extremely insightful for governments to understand their own and other countries’ strategies in terms of foreign policy and how these can evolve through time. To make this possible, you would have to build in highly accurate rationality and irrationality components into the agents’ behaviours. Such a powerful tool for foreign policy could perhaps aid countries to reach prosperous equilibrium states without the need for devastating armed conflicts.