I am a researcher working at the intersection of artificial intelligence and seismology. At ETH Zürich, I develop machine learning methods for seismic waveform analysis and earthquake physics, with the goal of building foundation models for seismology that are both scientifically grounded and useful for uncovering fault-zone processes and earthquake behavior.
My work spans earthquake forecasting across laboratory and natural settings, seismic denoising, and interpretable AI methods for geoscience. I am particularly interested in turning complex natural signals into meaningful physical insight.
I received my PhD in Data Science (cum laude) from Sapienza University of Rome, where my thesis focused on applying AI to seismology and earthquake physics. During my PhD, I conducted research at Los Alamos National Laboratory. I also hold a Master's degree in Data Science from Sapienza University of Rome and a Bachelor's degree in Sciences and Technologies for Media from University of Tor Vergata.
In addition to research, I have taught Machine Learning for Earth and Planetary Sciences at ETH Zürich and image processing at Tor Vergata University of Rome, with a focus on connecting theoretical concepts to practical applications.
My research focuses on machine learning for geoscience, seismic signal processing, foundation models for seismology, and interpretable AI for earthquake science. I am particularly interested in developing methods that connect data-driven models with physical understanding, from waveform representation learning and denoising to earthquake forecasting and fault-zone dynamics.
I work across a range of settings, from controlled laboratory experiments to large-scale Earth observations, aiming to bridge different spatial and temporal scales and to develop models that generalize across them.
More broadly, I care about robust and transferable approaches that operate across tasks, datasets, and scales—moving from task-specific solutions toward more unified frameworks for Earth science. I am especially motivated by using AI to extract meaningful structure from complex natural systems and to support scientific discovery.