My research work comprises two topics: forecasting methods and AI applied to the energy transition. I pursue the former on my daily job as an applied scientist at Amazon. For the second, I pursue my own research and I co-supervise a PhD student.

For both research topics, I am open to interesting collaborations and/or co-supervision of PhD students.


It is general wisdom that knowing what to expect, a.k.a. forecasting future events, is important to take better decisions and to reduce potential risks. For that reason, I like working on forecasting research as I believe forecasting algorithms can help solve some societal problems by improving decision making.

In particular, forecasting methods have always been part of my research interests. During my PhD, a large part of my research involved forecasting algorithms in the context of the energy transition. My research in this area received four awards: an award in a international competition organized by the French TSO, an award in a forecasting competition organized by a British TSO, an award for the most cited paper in the journal of Applied Energy, and the Cum Laude recognition upon my PhD graduation at TU Delft.

Currently, I pursue my interest into forecasting research with my role at Amazon, where my work involves researching on novel and state-of-the-art forecasting algorithms.

In my free time, I also pursue my own research interest in forecasting electricity prices. In that context, I am the developer of the epftoolbox python library, the first open-access library for driving research in electricity price forecasting.

AI for the energy transition

I believe the energy transition, i.e. the shift from carbon-based fuels to renewable sources, is one of the several societal changes required to tackle climate change. For that reason, a second topic that I am motivated about and that I work on is AI methods applied to the energy transition. In particular, I think that machine learning and AI have the potential to solve several of the existing problems associated with the energy transition, e.g. the issues associated with an increasing share of renewable sources in the electrical grid.

As part of my PhD research, I already spent four years working on machine learning, optimization, and other data science algorithms that could help advance the energy transition. Among others, I worked on forecasting energy prices and generation of renewables, optimal control of energy systems, and fault detection in smart grids.

At the moment, I combine my main role as an applied scientist at Amazon with my personal research interest into this topic. In particular, in my free time, I co-supervise a PhD student (Lola Botman) whose PhD research involves AI techniques that help advance the energy transition. In addition, I pursue my own research interest in forecasting electricity prices.