I chose the UvA MSc Computational Science programme primarily because of its emphasis on interdisciplinary study and the opportunity to work with leaders in the field such as Michael Lees, Alfons Hoekstra and Peter Sloot, to name a few. I chose to study computational science in general because solid skills in computation and simulation are becoming increasingly critical in almost all scientific fields.
My thesis focused on modelling heterogeneous traffic, which is traffic where there is a wide range of different vehicle types on the road (such as is found in places like India, Thailand, etc), and their interactions are qualitatively different than the well-studied homogeneous traffic flows which consist mainly of one or two vehicle types following very well-defined traffic rules. I chose this area because the existing research was relatively sparse (despite the majority of the world’s population living in places where heterogeneous traffic prevails), and it was a subject ripe for computational modelling because of the inherent difficulties with performing real-world traffic experiments.
My time in the Master’s was a great all-round experience. The freedom to choose many of my own courses was a big advantage, allowing me to gain a breadth of knowledge across many disciplines. I particularly enjoyed the courses Stochastic Simulation taught by Alfons Hoekstra and Complex Systems Simulation run by Michael Lees. They are both highly effective communicators at the forefront of their respective fields and their teaching is imbued with the passion that they have for their subjects. Most importantly, these two topics are fundamental to computational science which makes a solid grounding in them all the more important.
The Faculty of Science at the UvA did a good job of accommodating me as an international student. The general atmosphere at the Science Park campus is that of a high-tech hub, with the modern architecture and spacious design evoking the feeling of an early Google or Facebook. It is certainly an exciting environment in which to be working.
My first job after graduating was as a data scientist at a global market research agency. Even though I had no experience in the field, having solid training in science and maths was what got me the job.
Currently I spend much of my time applying techniques from natural language processing and machine learning to data from social media in order to understand and quantify public feeling towards brands and other entities. The work is a good mix of theory—for example, reading papers on the latest advances in these fields—and programming, both of which I enjoy.
While there is great potential for computational methods to be applied in the market research industry, I might eventually want to move back towards more traditional scientific modelling. The great thing about computational science is that the techniques can be applied in almost all domains.