Our first AI course in English with the European credential. Twelve editions already delivered in Italy: first English cohort now forming. Contact us for the dates
First AI course in English: twelve editions in ItalyLeonardo Cella
Member of AIPIA’s Technical-Scientific Committee · Quantitative Power Trader & Machine Learning Specialist · Milan
Leonardo Cella is a quantitative power trader specialising in machine learning, statistics and computational methods. He holds a PhD in Computer Science from the University of Milan and an MSc in Computer Science and Engineering from Politecnico di Milano, with a specialisation in machine learning.
He is based in Milan. As Lead Quantitative Power Trader he heads the quantitative desk at Duferco Energia, after a path running through academic research, data science applied to energy markets, and a period at Amazon as an Applied Scientist.
What he brings to the Committee
On the Technical-Scientific Committee, Leonardo Cella represents the meeting point between academic-grade research and systems running in production under real constraints of time and risk. It is a useful combination when the Committee has to tell a solid result from one that only holds on the test data.
He also brings experience from a sector, energy markets, in which machine learning models make continuous operational decisions and are judged on their results: a demanding test bed for the good practice the Committee develops.
Quantitative methods and artificial intelligence in energy markets
At Duferco Energia, since 2024, he has built from scratch an automated algorithmic trading system, continuously available and low latency, for the Italian continuous intraday market, and has led the definition and implementation of its strategies. He extended the setup developed for Italy to several European markets and led the development of an in-house machine-learning model for wind generation forecasting.
He leads artificial intelligence and machine learning initiatives within the trading division and built the quantitative team, selecting and coordinating its analysts. Before that, at Axpo Italia, he was Senior Data Scientist working on portfolio forecasting and on machine learning applications to imbalance trading.
Bandits, meta-learning and sequential learning
His doctorate, defended at the University of Milan with a thesis on efficiency and realism in stochastic bandits, focused on reinforcement learning. From 2020 to 2022 he was a postdoctoral researcher in the Computational Statistics and Machine Learning group at the Istituto Italiano di Tecnologia, working on non-stationary bandits and meta-learning in collaboration with researchers from META, Google and École Polytechnique.
The work from that period was published at the main machine learning venues: NeurIPS, AISTATS and ICML. Earlier, as an Applied Scientist at Amazon, he had developed a sequencing algorithm for music stations.
On the Technical-Scientific Committee
Appointment effective from 21 September 2026.
Join AIPIA
Membership gives access to the members' network, to continuing education and to the association's activities.