For more than twenty years my job has been to listen to a machine that hears almost everything.
A gravitational-wave detector is the most sensitive instrument ever built, and that sensitivity is the whole problem. It records two black holes colliding a billion light years away. It also records a truck on the road outside, a winter storm in the Atlantic, a screw slowly loosening inside a vacuum chamber. Almost everything that comes out of it is noise. My work is knowing the difference.
The work
The formal name for it is noise analysis and system identification. In practice it means this: before you can look for a signal, you have to make the noise stop shouting. You measure the spectrum of the instrument, you flatten it — whitening, we call it — and then you look at what is left standing. Sometimes what is left is a glitch, one of the thousands of small malfunctions a detector produces every day. Very rarely, it is the universe.
I build the algorithms that make that distinction: data conditioning, detector characterisation, transient classification. For the last decade much of it has been machine learning, which I started pushing into this field when the field still considered it a slightly disreputable idea. It is standard now. I am glad to have argued for it early.
How I got here
I grew up looking at a very dark sky and trying to picture what was behind it. It is the reason I became a physicist.
I studied in Pisa and wrote a thesis on the statistical analysis of the large-scale structure of the universe — the biggest thing there is to measure. Then I won a PhD fellowship in the Physics Department there, joined the Virgo experiment, and never really left. From 2008 to 2014 I led the noise analysis group. In September 2015 the detectors heard two black holes merge, and everything the collaboration had been preparing for over thirty years arrived in a fraction of a second.
I was in my office. The email arrived, the same one everyone in the collaboration received, and within minutes I was on Skype with colleagues. For thirty years the field had been waiting for that message, and the first thing I needed when it came was to say it out loud to people who knew exactly what it meant.
Teaching and outreach
I am a full professor at the University of Bologna, where I teach gravitational-wave physics. Before that I ran the Data Science Office at the European Gravitational Observatory, and I was associate faculty at the Scuola Normale Superiore.
I also spend a good part of the year in front of people who did not study physics: students, teachers, journalists, and once a TEDx stage in Putignano. A discovery that nobody outside the collaboration can picture is only half a discovery.
Photography and travel
The rest of this site is the rest of my life. I travel constantly for work, and I photograph while I do it. No commissions and no clients: a photographer for passion, for about as long as I have been a physicist.
I have come to think it is the same instinct as the day job. Both are about paying attention to something faint, and being patient enough to wait for it to show up.
In brief
- 2026 – Coordinator for Data Analysis and AI, CAOS Laboratory, Perugia
- 2024 – Full Professor, University of Bologna
- 2021 – 2025 Co-chair, Data Analysis Division, Einstein Telescope
- 2018 – 2024 Head of the Data Science Office, European Gravitational Observatory
- 2017 Breakthrough Prize in Fundamental Physics, with the LIGO–Virgo team
- 2018 – 2023 Proposer and chair, COST Action on gravitational waves, geophysics and machine learning
- 2014 – 2018 Scientific coordinator, GraWIToN training network
- 2008 – 2014 Leader of the noise analysis group, Virgo
- Co-founder and board member, EuCAIF, the European coalition for AI in fundamental physics
Publications, the full academic record and the CV live on my university pages. There is also a Wikipedia entry, kept up to date by someone who is not me. This site is for everything else.
For talks, collaborations or student supervision, write to me.