Simone Amoroso

Led a 50-person group developing Python data analyses for peer-reviewed publications using advanced statistical and machine learning techniques. Introduced and optimized Deep Neural Network algorithms with TensorFlow for Monte Carlo simulation parameter tuning, reducing simulation needs by a factor of five. Trained generative AI models in PyTorch to reproduce expensive simulations. Spearheaded development of open-source C++ code for non-linear regression. Lectured in particle physics and data analysis courses and mentored over 20 students from B.Sc. to PostDoc level.

Led development of the first CE-certified LLM-powered medical device, Prof. Valmed. Designed cloud-native MLOps infrastructure for ENBW’s energy trading analytics enabling scalable deployment and monitoring of predictive models. Architected testing and validation frameworks for AI/ML systems to ensure compliance and robustness in regulated applications. Delivered professional training on AI testing, EU regulatory frameworks, and quality assurance for production AI systems.

Led statistical combination of W-boson mass measurements, developing Python tools for uncertainty and correlation studies. As head of the Physics Modeling Group, managed cloud and HPC computing resources (~300K CPU cores) and defined long-term computing strategies (CPU and GPU). Led a team of 150 to commission, validate, and deploy novel theoretical models and Monte Carlo simulations in production.