About me

I am an assistant professor in the data science group at the Institute for Computing and Information Sciences, Radboud University Nijmegen.

My research focuses on probabilistic machine learning, with particular interest in uncertainty quantification, time series, and spatiotemporal data. I develop methods for reliable prediction in complex real-world systems, with applications in energy systems and weather and climate. Methodologically, my work draws on Gaussian processes, Bayesian inference, state-space models, probabilistic deep learning, and deep generative models. I am particularly motivated by models that are scientifically rigorous and also make it into operational use, supporting real decisions under uncertainty.

Previously I was a postdoc with prof. Tom Heskes at iCIS, and I hold a PhD from Maastricht University (“Bayesian Inference in Multivariate Nonlinear State-Space Models”), supervised by dr. Michael Eichler. My work on Gaussian processes began in prof. Magnus Rattray’s group in Manchester.

I currently coordinate SHARE, an NWO project on physics-informed, privacy-preserving synthetic data for the Dutch energy transition, and lead the AI and data science theme in the Radboud–Alliander partnership. See Research for my research themes and publications, or the full list on Google Scholar.