 Pesto is a novel geometric transformer that uses element labels to predict protein-protein interfaces. It outperforms existing methods in terms of accuracy and efficiency, making it suitable for large-scale analysis of protein structure data. This method has the potential to revolutionize our understanding of protein-protein interactions, enabling us to discover previously unknown biomolecular interactions and gain insights into the mechanisms behind them. This article was authored by Lucian F. Krap, Luciano A. Abriata, Fabio Cortes Rodriguez, and others.