I am an ELLIS PhD student at the University of Edinburgh, supervised by Antonio Vergari, Nikolay Malkin and Vincent Fortuin. I’m happy to be part of the APRIL lab at the University of Edinburgh and the ELPIS lab at Helmholtz AI and TU Munich.

I’m broadly interested in approximate Bayesian inference and uncertainty quantification. Currently, I am mostly working on variational inference (VI), aiming to derive formal guarantees for when VI will successfully recover certain characteristics of the target distribution.

If you are excited about similar topics, reach out - I am always happy to chat!

News

I’m very happy to present our poster on unlocking subtractive mixture models as variational families and proposals for importance sampling at AISTATS 2026 and give a contributed talk at the OPTIMAL@AISTATS workshop on robust VI with location-scale families.