About the author
From Physics to AI.
I am interested in how artificial intelligence can be used to address some of humanity's remaining challenges across fields such as energy, health, and materials science.
Current focus
Building scalable generative models for scientific applications. Currently developing scalable architectures for flow matching that can efficiently average over millions of training points during inference.
Working hypotheses
- • Scalable generative models require rethinking how we compute velocity fields.
- • Can we move from weight-based encoding to efficient averaging over massive training sets?
Timeline
Master of Science in Artificial Intelligence
University of Amsterdam 🇳🇱
Bachelor of Science in Physics Engineering
Instituto Superior Técnico 🇵🇹