I work on problems where data, software, and consequential decisions meet.
I'm an economist by training, with an MA in Economics from the University of Toronto. My work has ranged from institutional real estate finance and investment to econometrics, data science, and software. The common thread is applied decision-making: using data, models, and judgment to understand complex problems and decide what to do.
I spent three years in institutional real estate investment and development, working on underwriting, financing, capital allocation, and strategic decisions across a billion-dollar development program. Today, I'm a replicator for the Journal of Political Economy, working across empirical research pipelines and computational workflows.
I also contribute to open-source econometrics and build projects across machine learning, quantitative research, and software.
Much of my work comes back to one question: given the evidence, what should we actually do?
Contact
The fastest way to reach me is email. I'm also on LinkedIn and GitHub.
Work
Senior Financial Analyst, Chard
Worked across underwriting, capital allocation, debt financing, forecasting, and portfolio decisions for a $1B real estate development program. I also built data and market-intelligence systems to make those decisions faster and more repeatable.
Data Scientist, Sciences Po
Built a geospatial dataset from more than 50 million GPS observations to study how London's bus network operates, including efficiency, asset utilization, spatial competition, and market power.
Replicator, Journal of Political Economy
Reproduce and audit empirical research across R, Stata, Python, Julia, and shell, checking statistical methods, outputs, dependencies, and whether the results can actually be reproduced.
Projects
Marketplace simulation
Simulated orders and couriers with matching, routing, ETA estimation, batching, experimentation, monitoring, and CI/CD.
Mars Water Search
Four AI rovers, limited budgets, one discovery prize. An experiment in search, incentives, and collaboration.
Open source
Contributions to econometric inference tooling and multi-armed bandit algorithms.
Research
The Role of Social Capital in Business Performance: An Empirical Analysis
Using nearly 14 million US businesses with double machine learning and quantile regression to study how social capital shapes firm outcomes.
Viral Voting: A study in the information diffusion effects of different social pressure treatments on voter turnout.
Can social pressure spread from one household to the next—and change who turns out to vote?
Deep Hedging with Non-Parametric Volatility Surfaces and Explicit Forecasting
Can a reinforcement learning agent use the shape of volatility to make better hedging decisions?
Index Funds’ Corporate Stewardship Behaviour
What happens to corporate stewardship when ownership is concentrated in the largest passive investors?