I build quantitative models that connect rigorous mathematical theory — stochastic calculus, PDEs, stochastic processes — with production-ready code. My work spans actuarial modeling, macroeconomic analysis, and quantitative finance, always with a focus on results that are statistically sound and implementable in practice.
Currently finishing my undergraduate degree in Applied Mathematics at UFRJ (June 2026) and open to freelance projects in quantitative modeling, risk analysis, time series forecasting, and data science for finance.
Itaú Quant Challenge 2025 — Top 4% (40/953)
A long-only equity strategy for the Brazilian market that detects market regimes and adapts portfolio allocation accordingly. Combines topological data analysis (persistent homology), factor overlays via Ridge/ElasticNet metamodels, and hierarchical risk parity (HRP) for robust allocation.
- Result: Sharpe ratio of 1.18 at 88.4% statistical confidence on out-of-sample Brazilian market data
- Stack: Python, Scikit-learn, Giotto-TDA, Riskfolio-Lib
Undergraduate Research — UFRJ
Models the strategic interaction between high-frequency traders and market makers using Mean Field Game theory. Solves the coupled HJB–Fokker-Planck PDE system numerically using finite difference methods (Lax-Friedrichs) and Picard iteration. Calibrated with real B3 historical data (1986–2025).
- Result: Mathematically demonstrated the emergence of liquidity resilience and liquidity crunch phenomena from agent interaction alone
- Stack: Python, NumPy, SciPy
- Quantitative Finance: alpha research, factor models, backtesting (vectorized, purged cross-validation), portfolio optimization (Markowitz, HRP), risk management
- Statistical Modeling: time series (ARIMA, GARCH), survival models, hypothesis testing, regression analysis
- Machine Learning: supervised/unsupervised learning, NLP, LLMs, RAG systems, deep learning
- Mathematical Foundations: stochastic calculus, PDEs, optimization, linear algebra, numerical methods
Open to freelance projects in quantitative modeling and data science — feel free to reach out via LinkedIn or email.



