About Me

I'm an Italian researcher and data scientist (master degree in Theorical Physics and PhD in Civil Engineering) with a long experience in research, analytical methods and data science.

In my career I have got the possibility to reinvent myself several times. I have been a professional in the field of Unix system administration for a couple of years, a web applications developer, a co-founder of a research laboratory, and recently an innovation officer for a startup that is rewriting the rules of data protection.

Back to science, I have been a researcher in the field of Statistics of extreme events and hydrology, then Complex networks with applications in Economics and Social Networks, and - more recently - a practitioner of applied machine learning. Along this path I lived in 3 different countries (Italy, Uk, and Switzerland) where I also had the possibility to teach in Universities (ETH Zurich, Univ. of Cagliari, LUISS, IMT High studies Lucca), and for many years in private education companies (Talent Garden). I tutored more than 200 students in several classes, both in English and Italian language, teaching introductory machine learning, natural language processing, python for numerical analysis and visualization, deep learning with TensorFlow, general data visualization, and network analysis. I am a faculty member of the Business Data Science master that the Talent Garden company organizes in Vienna and Madrid, for them I am in charge of the modules of machine learning. In these years, I used all sorts of e-learning tools from Moodle, to flipbooks and dedicated platforms. To teach Python and coding I extensively use the Jupyter Notebook.

There is a red line connecting every scientific work I wrote these years: the creation of a specialized databases. I have built datasets of public finance (balance sheet of Italian municipalities), datasets of financial statements to study systemic risk, databases of climatic data with geolocation, and many databases of tweets. I used tweets to find weak signals of pneumonia symptoms during the initial diffusion of the Covid-19 pandemics (an article that published in Scientific Reports that attracted a lot of attention from the media), or conversely, large trends of opinions and supporters in political elections. I published these multidisciplinary researches in journals like Scientific Reports, Plos One, and in a more economic side, Health Policy, Structural Change and Economic dynamics, Metroeconomica, and Social Science \& Medicine. I have been cited more than 1800 times, with an h-index of 15 with the majority of this scientific production occurring after 2011,the year of my PhD.

The most successful research paper was cited for more than 800 times (DebtRank too central to fail) while another one got a lot of attention from the International Press (Early warnings of Covid-19 in Twitter was briefly cited by Financial Times, and got a space in more than 66 news source in different languages source).

Looking at the technology side, I consider myself a good programmer, I am using mostly Python for my everyday activities of data analysis and problem solving. But I also spent a lot of time using R, and a long time ago, Perl, C, and even FORTRAN. Along with SAS and Matlab I have encompassed the entire spectrum of modern Data Science.

While a Data scientist must be good at coding, he must also be competitive in the realm of algorithmic thinking, and analytical design. Personally, I believe that statistical inference,testing and hypothesis validation, and Monte Carlo methods, are the “solution” to many of the problems a Data Scientist can encounter, and I try to apply these methods whenever possible. In conclusion I am ready to reinvent myself once again.