I'm a computer engineer turned quantitative finance researcher. After 6 years of blockchain and backend development (Node.js, TypeScript), I'm now pursuing an MSc in Sustainable Finance & Data Analytics at LUMSA University in Rome — focused on machine learning for asset pricing and financial econometrics.
My thesis reproduces Bryan Kelly, Semyon Malamud, and Kangying Zhou (2024) — "The Virtue of Complexity in Return Prediction", exploring overparameterized models, ridgeless regression, and Random Fourier Features for equity return forecasting, then applies this methodology to Loss Given Default (LGD) prediction in credit risk.
- Overparameterized ML models & the double descent phenomenon in finance
- Ridgeless regression, Random Fourier Features, and high-dimensional asset pricing
- Credit risk modeling (LGD forecasting)
- Building reproducible research tools in Python
Current (Research & Data Science)
Python | NumPy | Pandas | Scikit-learn | Statsmodels | Matplotlib | LaTeX
Previous (Backend & Blockchain)
TypeScript | Node.js | NestJS | PostgreSQL | MongoDB | AWS | Solidity | Web3.js
Open to data science, data analytics, quantitative research, and economics research roles in Rome, Turin, remote, or across Europe.
- English — Professional
- Turkish — Native
- Italian — Learning (A2)