Assistant Professor in Finance, HEC Montréal (from August 2026)
Ph.D. in Economics and Finance, Università Bocconi (2026)
Thesis: Reinforcement Learning Approach to Continuous Mean-Variance Portfolio Selection.
Thesis: Around the Validity of International Assessments on Mathematics during Obligatory School.
Thesis: The Geometry of Tessellations. The (2,3,7)-tessellation.
I develop a multi-agent LLM framework that processes Federal Reserve communications to construct narrative monetary policy surprises. By analyzing Beige Books and Minutes released before each FOMC meeting, the system generates conditional expectations that yield less noisy surprises than market-based measures. These surprises produce theoretically consistent impulse responses where contractionary shocks generate persistent disinflationary effects, and enable profitable yield curve trading strategies that outperform alternatives. By directly extracting expectations rather than cleaning surprises ex post, this approach demonstrates how multi-agent LLMs can implement narrative identification at scale without contamination in high-frequency measures.
We propose a progress-oriented score that improves capital allocation when firms have valuable environmental growth options and CEOs hold private information about quality. Assessing decarbonization plans separates good (green) from bad (brown) options and yields superior investment performance.
We introduce a network-based methodology to measure emissions along complex international supply chains, enabling counterfactual experiments and forecasts for high-scope emissions under scenarios like maritime disruption, conflicts, trade wars, and revised carbon taxes.
We study stochastic-trend monetary policy rules linking potential output growth, demographics, and inflation expectations to prevailing rate trends. The model improves US short-rate forecasts; for the Eurozone, cautious policy following the US proves more effective than fragmentation-focused rules.
We show that aligning term-structure models with salient data features improves forecasts of US short rates and yields stationary term premia, highlighting the benefits of data-congruent specifications.
We model networks with emission externalities where greenness shapes borrowing constraints. A DAO that allocates capital internally and records emissions on-chain replicates the first-best; preliminary evidence suggests sizeable welfare gains.
PhD short course on using large language models as tools for economic research.
Advanced: Python, Matlab, C++, LaTeX
Intermediate: R, RStudio, Git, GitHub, AWS
Basic: HTML, CSS, JavaScript
Spanish (Native) · English (Advanced) · Italian (Advanced)