Research

My research focuses on developing machine learning methods for molecular sciences, with applications ranging from drug discovery to materials design. The work spans quantum chemistry, neural network potentials, generative models, and high-throughput computational screening.

Research Philosophy

We believe that machine learning will fundamentally transform how chemistry is practiced. Our goal is to develop methods that are not just accurate, but also interpretable, transferable, and practically useful for real-world chemical problems.

Our approach combines rigorous physical foundations with modern deep learning techniques. By encoding physical laws and chemical knowledge into neural network architectures, we create models that are both data-efficient and physically meaningful.

Collaboration

We actively collaborate with experimental chemists, pharmaceutical companies, and national laboratories. Our open-source tools are used by researchers worldwide and have been integrated into several commercial platforms.

If you're interested in collaboration opportunities, please visit the contact page for more information.

Impact

158
publications
18,400
citations
h-58
h-index
$15M+
total funding