I work on how uncertainty is represented and carried through systems, from measurement to decision. My research develops probabilistic methods that keep a system's uncertainty sound as it propagates: scalable variational inference, Bayesian state estimation on manifolds, and proper-scoring diagnostics for whether a model's confidence is warranted.

I am a DDSA Fellow and postdoctoral researcher in the SQUARE group at the IT University of Copenhagen. I completed my PhD with Thomas Hamelryck at the University of Copenhagen.

Current work spans real-time Bayesian filtering for pose estimation, sonar-based SLAM, and theory for mixture flows in variational filtering.