Researcher at the intersection of scientific machine learning and physics-based modeling, focused on uncertainty quantification, differentiable programming, and physics-informed neural networks (PINNs). Currently a PhD student at the GATE Institute, where I develop interval PINNs for structural dynamics and work on surrogate models for engineering applications. I also contribute to the MODE and TAMBO Collaborations, building fully differentiable surrogate pipelines for particle shower simulation and detector array optimisation. As a Data Science Consultant at Adastra, I build cloud-based ML solutions on AWS, including RAG systems and optimisers. My background spans electromagnetics research at KTH and Sofia University, as well as human electromagnetic exposure studies with Ericsson AB on 6G wireless systems. I hold an MSc in Electromagnetics, Fusion and Space Engineering from KTH and a BSc in Information Technology/Electrical Engineering from TU Vienna. I'm passionate about bridging rigorous scientific modeling with modern ML tools to solve problems in physics, engineering, and applied mathematics.