I am a staff research scientist at Spotify. My main motivation of research is to develop machine learning systems that can automatically learn from large amount of unlabeled data and make informed decisions. In particular, my main areas of research are: (1) probabilistic modeling, especially latent variable models and Gaussian processes; (2) scalable inference methods for probabilistic models such as stochastic variational inference; (3) automated decision-making methods such as Bayesian optimization, automated machine learning.
Prior to joining Spotify, I was a machine learning scientist at Amazon Research Cambridge, working with Neil Lawrence. Previously, I worked as a co-founder at our startup, Interentia Limited, together with Neil Lawrence, Andreas Damianou and Javier Gonzalez, which was later acqui-hired by Amazon. Before that, I worked as a postdoctoral researcher at the university of Sheffield with Neil Lawrence, after completing my PhD degree with Jörg Lücke on machine learning at the Goethe University Frankfurt.