I am a machine learning researcher at Twitter, as well as PhD student at Imperial College London supervised by Prof. Michael Bronstein. My main research interest lies is on combining structure and time in machine learning models. Data with both a structural and a temporal component can be often represented as graphs which evolve over time (dynamic graphs), with examples such social networks, the spread of a virus in a network, financial transaction networks or video scene graphs. The temporal dimension of such graphs often contains critical information which goes otherwise lost when considering only a static graph, which is however what has mostly been done in Graph Neural Networks (GNNs) research so far. Moreover, I’m interested in making GNNs scale to very large graphs, since most interesting real-world applications (including the ones we work on at Twitter) have graphs with up to billions of nodes and edges.

Previous to my current position, I was working at Fabula AI, which was then acquired by Twitter in June 2019.

In June 2019, I graduated with a distinction from an MPhil in Advanced Computer Science at Cambridge University. My thesis was on graph deep learning models on RNA data, under the supervision of Prof. Pietro Liò and Prof. Michael Bronstein.

In June 2018, I graduated with a BEng in Computer Science from Imperial College London. During my time there, I interned in 2017 at G-Research and in 2018 at Google as a software engineer. During summer 2018, after my graduation, I worked with Prof. Michael Bronstein on fake news detection using graph convolutional networks.

I am also the co-founder of LeadTheFuture, a non-profit organization with the goal of helping talented Italian students in STEM to achieve their potential. Today LeadTheFuture counts more than 100 mentors (engineers, researchers and entrepreneurs) and 500 selected mentees.