��ɫ��Ƶ

Our daily lives are shaped by atom-scale processes, from drug molecules binding to proteins to lithium ions diffusing in phone batteries. Experimentation at this scale under realistic conditions is challenging, but our understanding of the underlying physics can act as a magnifying glass with atomic resolution. Accurate descriptions of atom scale interactions require a quantum mechanical treatment. However, direct quantum mechanical calculations are computationally expensive, making this unfeasible for large systems. Lars explores the use of machine learning to accelerate simulations from years to minutes. Traditionally, new materials have been discovered through trial and error. With accelerated atomistic simulations we can screen candidate materials and drugs before synthesis, speeding up the discovery process.

Originally, Lars studied Theoretical Physics at the University of Birmingham, with a focus on Astrophysics. During his internship at the Max Plank Institute for Nuclear Physics, Lars made his first contact with scientific computing while working on a high-energy camera that is set to observe x-rays emitted by cosmic particle accelerators. Changing to the University of Cambridge for his Master's, Lars started focusing on condensed matter physics. Here he discovered his passion for computational modelling at the atomic scale.

During his PhD Lars worked on using machine learning accelerated simulations to understand catalytic reactions. The ultimate aim is to find new materials which can accelerate reactions to reduce global energy consumption. In addition to his applied work, Lars contributed to developing novel machine learning architectures designed to capture non-local interactions.