I am a research scientist at the Max Planck Institute for Intelligent Systems, with a background in physics and computer science from Cambridge. I am interested in the mathematical and computational foundations of intelligence, in particular how optimisation, architecture, data, and objectives shape learning dynamics, generalisation, and emergent behaviours such as transfer learning and in-context learning. Through my research, I seek to understand the inductive biases that shape different solutions and how to design novel interventions that explicitly target desired behaviours.
I am currently working with Antonio Orvieto on discrete diffusion models for biological sequence data, aiming to learn meaningful representations for genome modelling.