PhD researcher working on continual learning with deep neural networks (i.e. how intelligent systems can keep learning and adapt). My work draws on statistical physics and dynamical systems, and extends to LLM training, (multi-agent) reinforcement learning, and AI safety.
Alongside my technical research, I think and speak about the (existential) risks of advanced AI, often linking AI to broader themes such as geopolitical instability and energy. I have been invited to speak at the Flemish Parliament, VRT (national television), universities, industry, etc., and I am happy to give talks on this topic and on AI in general to help raise awareness and improve public AI literacy.
Other interests include effective altruism, game theory, AI for science (e.g. protein folding), nuclear fusion, cosmology, the evolution of cooperation, longevity, mindfulness, climbing, and trail running.
Text box item sample content