Tommaso Marchetti

I am a Fixed-term Researcher (RTD-A) in Computer Science at Università Cattolica del Sacro Cuore, Rome, Italy.

My research lives at the intersection of machine learning, statistical inference, and the analysis of complex scientific and biomedical data. I am especially interested in problems where the data are heterogeneous, multimodal and not abundant, and where physical, biological or astronomical structure has to be combined with modern AI techniques to produce models that are accurate, interpretable, and operationally useful.

What I currently work on

  - AI / machine learning for biomedicine: deep learning, large language models and multi-agent systems on heterogeneous clinical and physiological data, including AI on neurophysiological signals (EEG) and on metabolic imaging.

  - Statistical inference and data mining on large heterogeneous catalogues: Bayesian inference, rare-event detection, and multi-survey cross-matching, originally developed in observational astronomy and applied across scientific domains.

  - Operational machine learning: high-precision predictive pipelines, active learning, and model validation under data scarcity and uncertainty.

Background

I obtained my PhD in Astronomy at Leiden Observatory in 2019, with a thesis on hypervelocity stars in the Milky Way ("Hunting for the Fastest Stars in the Milky Way"). I subsequently worked as an ESO Fellow at the European Southern Observatory in Garching, Germany (2019–2022); as a Machine Learning Scientist at ignitia, Sweden (2024–2025), on high-precision tropical rainfall prediction and thunderstorm nowcasting; and as a Data Scientist at Fondazione Policlinico Universitario Agostino Gemelli IRCCS in Rome (2025–2026), on AI for clinical data analysis.

Outside research, I am volunteering with Baobab Experience, a humanitarian non-profit supporting migrant persons in Italy and Europe. 

In my spare time I read, travel, and play bass, saxophone and synthesisers.