— Italo Calvino
Hi there. I am a research scientist specialising in AI-driven simulation methods for materials discovery.
I combine first-principles techniques with machine learning to design high-throughput simulation workflows for materials at scale. I have defined the scientific strategy and co-led the technical execution of large-scale international researchs collaborations, working closely with experimental and engineering teams to translate research into applied outcomes. I've spent over five years at startups focused on materials science, sitting at the intersection where cutting edge AI research meets engineering.
My core expertise lies in ab-initio methods to predict the electronic (DFT, DMFT), structure (MLIPs, structure searching, generative design, phonons), and transport properties of complex materials systems critical for future energy technologies, including semiconductors, transition-metal oxides, battery materials, and spin defects.
I develop AI methods to accelerate and enhance materials simulation. A central tenet is physics-informed machine learning — I've designed enhanced exchange-correlation functionals and created the first DMFT impurity solver using delta-learning. I integrate these with workflow automation (AiiDA), high-throughput screening (pymatgen), and machine-learning interatomic potentials to enable materials discovery and exploration at scale.
I'm a passionate hill walker, staunch advocate of the slow food movement, and always keen to learn something new.
