Particle physics × machine learning
I’m a PhD researcher at ETH Zurich, based at CERN. I develop machine-learning methods for high-energy neutrino event reconstruction as a member of the FASER Collaboration.
01
My work combines experimental particle physics with machine learning, focusing on sparse neutrino-detector data, self-supervised representations, and efficient reconstruction.
02
Peer-reviewed work and recent preprints.
Featured work
A sparse vision-transformer framework that learns transferable representations from heterogeneous detector data, improving multiple reconstruction tasks while reducing dependence on labelled events.
Spiking neural networks learn to identify charged-particle trajectories in a CMS-like detector without supervision, even in the presence of substantial noise.
A validated reconstruction and identification framework for electromagnetic showers in the FASERν emulsion detector using CERN SPS test-beam data.
03
Recent presentations on machine learning for particle detectors.
04
2025 — present
Current position
ETH Zurich · based at CERN · FASER Collaboration
Developing machine-learning methods for TeV-scale neutrino event reconstruction, with a focus on sparse detector data, self-supervised pre-training, and reusable foundation-style models.
2023 — 2025
IMAPP joint European programme
Université Clermont Auvergne · TU Dortmund University · University of Bologna. Graduated cum laude, with training across theory, detector physics, data analysis, and scientific computing.
Summer 2024
CERN · CMS Experiment
Developed deep-learning approaches for Level-1 barrel-muon reconstruction, targeting low-latency FPGA deployment.
2024
Graduated with highest honours · Très Bien
Advanced training in statistics, deep learning, PyTorch, clustering, and scientific Python.
2020 — 2023
University of Padua
Bachelor’s thesis on machine-learning selection of inverse beta decay events for the JUNO experiment.
2019
Liceo Scientifico Pier Paolo Pasolini
Final grade: 100/100.
Contact
For collaborations, research questions, or speaking invitations, the best way to reach me is by email.
fcufino@ethz.ch