My research focuses on accelerating the discovery and understanding of magnetic materials through geometric deep learning. With a Master's in Physics, a Bachelor's in Computer Science, and over six years of professional software engineering experience, I bring strong computational rigor and production-level coding standards to complex many-body physics problems.
Currently, my work centers on developing a custom Equivariant Graph Neural Network (DSpinGNN) to simulate spin-lattice dynamics and magnetization coupling in 2D materials like CrI3. By leveraging PyTorch, the e3nn library, and first-principles data generated via Density Functional Theory (Quantum Espresso), I build state-of-the-art models from scratch to predict critical thermal properties and uncover renormalized Tc temperatures.
Beyond machine learning, I am deeply invested in building robust scientific tooling. I independently developed Phystrackx, a proprietary video-tracking software designed to streamline data acquisition in experimental physics labs. I am driven by the challenge of translating physical symmetries into efficient code and pushing the boundaries of what data-driven methods can achieve in statistical mechanics.
- ⚡ In my free time, I love listening to audiobooks, contemplating on investing, and going on long calm road drives.
