Senior Data Scientist · Machine Learning Researcher · Ph.D.
Holds a degree in Civil Engineering from the Polytechnic School of the University of São Paulo (USP) and was an undergraduate researcher in a FINEP project. Holds a Master's degree in Economics from Fundação Getúlio Vargas (FGV), focusing on the topological optimization of artificial neural networks for financial asset price forecasting.
Previously worked as an analyst at the Itaú-Unibanco Group and as a Product Manager at Itaú BBA. Holds a Ph.D. in Computer Engineering from the University of São Paulo (Escola Politécnica) and a Ph.D. in Applied Mathematics from the University of Turin and the Polytechnic University of Turin.
Research interests include Reinforcement Learning applied to financial decision-making processes (automated trading, portfolio optimization, and option pricing and execution) and Operations Research (production and logistics optimization). Currently works as a Data Scientist and Postdoctoral Researcher at the Aeronautics Institute of Technology (ITA), focusing on Genetic Programming, Metaheuristics, and Hyper-Heuristics for optimization problems.
All the professional roles I have held, both nationally and internationally, across corporate and academic sectors.
Academic journey across applied mathematics, economics, and engineering.
Exploring reinforcement learning and dynamic programming for perishable inventory decisions.
Reinforcement learning for discrete lot-sizing and stochastic optimization problems.
Machine learning for finance with emphasis on supervised and reinforcement learning approaches.
Advanced econometrics and policy analysis supporting data-driven financial decisions.
Quantitative foundations and systems thinking applied to engineering and analytics.
Selected publications covering reinforcement learning, optimization, and forecasting.
Reinforcement learning method for stochastic environments emphasizing post-decision states.
Hybrid dynamic programming and RL approach to discrete lot-sizing on parallel machines.
Linear discrete choice models for perishable products in stochastic demand settings.
Comprehensive survey of RL strategies for algorithmic trading.
Stacked neural architectures outperforming baselines in financial time series prediction.
Reinforcement learning strategy for exercising financial options under uncertainty.
A supervised framework that surpasses reinforcement learning traders in cryptocurrency markets.
Stock-out substitution modeling for vertically differentiated perishables.
Conference contribution blending investor sentiment with reinforcement learning for adaptive trading agents.
WaveNet, RNN, and ML baselines evaluated for Bitcoin price prediction.
Study of neural networks applied to forecasting Brazilian stock returns.
Projects
Open-source initiatives translating research insights into reusable tooling.
Live activity across public repositories.
Simulation environment and heuristics for managing perishable inventory with stochastic demand.
Batch utility to rename PDF files using pattern matching and metadata extraction.
Lightweight translation interface leveraging modern NLP services for research teams.
Post-decision proximal policy optimization experiments featuring dual-critic architectures.
Contextual bandit framework combined with ResNet models for adaptive trading strategies.
Reinforcement learning agents tackling discrete lot-sizing on parallel machines.
Convolutional models to approximate optimal stopping boundaries for financial options.
Exploration of Dempster-Shafer theory for evidence aggregation in uncertain systems.
Reach out for research partnerships, data science initiatives, or speaking engagements.