Research
Sequential decision-making under uncertainty: time-series forecasting, reinforcement learning, stochastic control with uncertain horizon, and the numerical methods that make them computable.
Projects
Signature methods for optimal control
in progress- Using path signatures (rough path theory) to learn approximate-optimal policies directly from time-series via REINFORCE.
- Connects modern ML with classical stochastic control.
Delegating portfolio management under time uncertainty
- Principal–agent theory with a Second Order BSDE (2BSDE) under default risk; new HJB PDE.
- Iterative Physics-Informed Neural Network (PINN) solver in TensorFlow.
- Best Paper Finalist (top 6) at SIAM Conference on Financial Mathematics 2025.
2BSDE with uncertain horizon and control in erratic environments
- Existence and uniqueness for a class of non-Markovian 2BSDEs with random horizon.
- Framework for robust control under both volatility and time uncertainty.
Predictive maintenance for rail systems — INFORMS RAS 2025
2nd place- With my team MathConvoy, we placed second at the INFORMS RAS 2025 Problem Solving Competition at the 2025 INFORMS Annual Meeting.
- Using operations research and advanced analytical approaches, we created a predictive maintenance tool for rail systems. Using wheel-profile data, mileage, railcar attributes and other sensor measurements, the tool predicts wheel failures on commercial trains.
- Joint work with my teammates Daniele Gioia, Jacopo Bonari, and Edoardo Fadda; the team code is shared in Daniele’s repository.
Linear decision rules for multi-stage stochastic programs (M.Sc. thesis)
- Tractable approximations of large-scale multi-stage stochastic programs via linear decision rules, applied to an assemble-to-order (ATO) problem.
- Advised by Prof. Brandimarte and Prof. Fadda (Politecnico di Torino).
- Repository cleaned and maintained by my collaborator Daniele Gioia.
Education
Aug 2021 – May 2026
UC Berkeley — Ph.D., Industrial Engineering & Operations Research
Collegio Carlo Alberto — M.Sc., Statistics & Applied Math
Politecnico di Torino — B.Sc. / M.Sc., Mathematical Engineering
Teaching
-
INDENG 241 Risk Modeling, Simulation, and Data Analysis
-
INDENG 222 Financial Engineering
-
MFE 230D Derivatives: Quantitative Methods
-
INDENG 243 Analytics Lab
-
INDENG 242 Applications in Data Analysis
-
INDENG 120 Principles of Engineering Economics
Talks & conferences
Invited speaker
- Bachelier FS26 — Bologna
- SIAM FM25 — Miami
- WCMF25 — USC, Los Angeles
Other
- Berkeley–Columbia Meeting in Engineering & Statistics — 2025
- INFORMS25 — RAS Problem Solving Competition (2nd place), 2025
- Berkeley CREST Cyber-Risk Conference — 2023
Awards
Jul 2025
Best Paper Finalist (top 6) — SIAM Conference on Financial Mathematics
Oct 2025
INFORMS RAS Problem Solving Competition — 2nd place
2025
Outstanding Graduate Student Instructor Award — UC Berkeley
Get in touch
(my name) dot (my last name) (at) (berkeley dot edu)
Best for collaboration, research questions, and CV requests.