I started my career in nuclear and energetic engineering and have a bachelors and master’s degree in engineering from the Polytechnic of Turin. I also have a diploma as a geometer, so I have always been deeply involved with mathematics throughout my academic and professional life. In my early jobs, I worked on data analysis and mathematical modeling of nuclear power plants. It was fascinating work, but then the disaster at the Fukushima plant in Japan happened and it became a bad time for nuclear power. So, I made a transition, staying within the energy industry, but doing evaluation of power plants, using both renewable and traditional energy sources. The work was part of a due diligence and development process for banks, looking at the facilities from technical, compliance, and financial points of view.  
 

This work led to my current role with eVISO S.p.A. - Algo Efficiency for Commodities, where I am the Director of Algo Intelligence. The company was a start-up at the time I joined; it has since completed an IPO and is now a public company. The challenge for this young firm was to open a brand-new energy retailer in the Italian market, which has historically been deregulated. We built this retailer from the ground up, providing energy to the final customer, with a focus on spot market. We used machine learning and other models to evaluate the energy consumption throughout the day and developed systems to trade proper quantities to meet demand. We eventually entered into a continuous trading regime and most recently have been facing the challenge of volatility in gas prices due to the energy crisis in Europe and the war in the Ukraine. In this situation, there have been difficulties, but also new opportunities opening up, particularly related to trading at higher frequencies than were customary in this market. 

At this point, I decided to pursue the CQF. I was very happy with my work and did not want to take time off for a PhD, but I was looking for something deeply practical and insightful to help our work the valuation of hedging instruments and derivatives as well as moving to new trading regimes. The entire CQF program was great, and for me modules three and four were the most useful, covering derivatives, especially options, and machine learning. In our day-to-day work, we run forecasts for every single point of consumption for every single client and so it was very helpful to dig deeper into Monte Carlo models and also to develop a solid foundation in Python. At work we had used Matlab, and R, and moved almost completely into Python in recent years, but this was very much a process of learning and building along the way. It’s very useful to take a step back and understand the more formal aspects of programming and data architecture to develop more robust systems.  

As a practitioner in the spot market, we are often working with an automated, systematic environment, combined with machine learning methods like classification trees and regression trees. So, in fact I was able to complete the final project applying theory and methods that are practical and genuine, not simply driven by fancy results. My plan now is to share with my team all these ways of implementing strategies and models, to be aware of the common pitfalls and best practices to follow in our day-to-day implementation, and also to explore how we will address the continuous power market over time. For anyone who is confronting real world problems and emerging opportunities in the financial or commodity markets, I would certainly recommend the CQF highly.  It takes a rigorous analytical approach and both the program structure and the assignments ensure that you develop a solid foundation across the essential areas of quant finance and programming in Python for these contexts.