CV

Data Scientist

Quantitative modelling and machine learning for high-dimensional and time-series data.

• Developed forecasting models using ARIMA/SARIMA, XGBoost and LSTM for revenue time-series prediction.
• Led unsupervised learning projects using HDBSCAN and dimensionality-reduction techniques to identify structure in complex project data.
• Co-designed an internal chatbot enhanced with a reinforcement-learning mechanism to optimise responses, managing state transitions and reward-based updates.
• Implemented complex SQL queries for large-scale data extraction and integration.

Research Intern

Developed a ML model to predict geopolitical risks with interactive visualization; applied XAI techniques to improve neural network interpretability.

MSc Stochastics and Data Science

Thesis: Efficient estimation of time-to-event models and quantile treatment effects under right-censoring. Developed and validated asymptotic properties of an extension of a causal inference method for right-censored data, providing a robust framework for estimating treatment effects in survival settings.

Supervisors: Matteo Giordano, Stijn Vansteelandt.

Coursework: Probability Theory, Statistical Inference, Stochastic Processes, Multivariate Statistical Analysis, Functional Analysis.

Exchange Semester (Erasmus+)

Coursework: Causal ML, Deep Learning, NLP.

BSc Mathematics for Finance and Insurance

Thesis: One-dimensional and two-dimensional random walk with extension to Skew walk and non-Markovian walks.

Supervisor: Cristina Zucca.

Coursework: Probability and Statistics, Numerical Analysis, Mathematics for Finance and Insurance, Linear Algebra, Analytical Methods.

Integrating XAI for Predictive Conflict Analytics

Macis, L.; Tagliapietra, M.; Castelnovo, A.; Regoli, D.; Greco, G.; Cosentini, A.; Pisano, P.; Carroccetto, E.</stro