Projects

A fully automated market-making engine I run on prediction markets: real-time order-book ingestion, automated quoting, and inventory control. Across two live accounts it has booked $65K+ in net P&L to date, over ~300,000 closed positions in ~18,000 distinct markets across ~900 series. The measured Jan–Jun 2026 window covers 287,920 of those positions, over which it recorded a 70.6% win rate, a 5.9 annualized Sharpe on daily dollar P&L, and a $473 maximum drawdown in cumulative P&L.

Solo projectRead more →

Co-founder and CTO of a Y Combinator–backed startup in San Francisco. Writing ~6,000 commits, ~90% of the team’s total, I built a Python ingestion pipeline, a provenance-tracked knowledge graph over ~140 PostgreSQL tables, and a TypeScript forecasting backend across a ~280K-line system. The product reached ~$60K in annual recurring revenue.

Co-founder & CTORead more →

A mobile social app I built and shipped (Flutter + Firebase) that grew to 500+ users across a few local high schools, with real-time feeds, anonymous posting, polls, and DMs. Launched with an App Store waiting-room trick and an iPad giveaway funded by Super Bowl betting promos.

Primary developerRead more →

I gave two frontier LLMs, GPT-4.1 and Terra, the same sunk-cost and lottery decisions as a chooser, a financial advisor, and a predictor of human behavior. GPT-4.1 split sharply by role; Terra did not. The awkward part: the advisor was not reliably rational.

Independent researchRead more →

My senior thesis. I designed an evaluation suite to test whether language models can predict how people answer survey questions, with normalized accuracy metrics, uniform and distributional baselines, and a seen/unseen split. A dummy-option ablation separated 'learning the format' from 'learning the behavior,' and turned up a negative result.

Senior thesis · advised by Dr. David van DijkRead more →

A study of how gradient-based optimizers hold up when a fraction of the training data is adversarially corrupted. I compared Robust Gradient Descent against SEVER, worked through the theory, and ran simulations across contamination levels, including a hybrid of the two.

Course research · CPSC 561Read more →Source →

A physics-research method for measuring superconducting energy gaps from photoemission data. Instead of fitting one energy slice at the Fermi momentum, a two-stage '1.5D' fit reads the gap off the whole band's dispersion, which cuts fit residuals several-fold (up to ~20×) when the gap is smaller than the instrument's resolution. 2nd-place poster at a Berkeley Advanced Light Source user meeting.

Yale Applied Physics · advised by Prof. Yu HeRead more →Source →

My girlfriend and I wanted a game we could play together over a call, and everything we found either wanted us both to buy a copy or ran on a matchmaking server that somebody will stop paying for eventually. So I built one that needs neither. Three escape rooms, two browser tabs, one rule: the thing you operate and the writing that tells you how to set it are never on the same screen. I designed the rooms; coding agents I directed wrote the 21K lines of JavaScript that implement them, in a day.

Design and directionRead more →Source →

Two people take turns crossing restaurants off a list until one is left. A short Monte Carlo essay on why the nightly dinner negotiation is secretly a mechanism-design problem, and why cutting the list by fractions beats cutting it by fixed numbers when nobody has the patience for more than a few rounds.

Read more →

Stores lean their mirrors back because it makes you look taller. Finding the best angle sounds like a one-line optimization until you try to pin down what "taller" even means. It turns out to be at least three different things, and they disagree. An optics puzzle that started in a Uniqlo fitting room in 2021.

Read more →