About
I'm Ian. I build ML systems and run experiments on how language models behave. Right now I direct LabOS, an agent-driven ML research lab. Before that I was co-founder and CTO of Axis, a Y Combinator–backed startup. I graduated from Yale with a B.S. in Computer Science in May 2025, where I focused on machine learning and statistics. (Magna cum laude, Phi Beta Kappa, Distinction in the Major.)
Now
I direct LabOS, an agent-driven research lab where AI agents run the experiments and I set the questions and read the results.
On the side, I design and run an algorithmic market-making system on Kalshi's prediction markets: order-book ingestion, automated quoting, and inventory control, with hard risk limits. Across two live accounts it has booked $65K+ in net P&L to date, over ~300,000 closed positions in ~18,000 distinct markets. In the measured Jan–Jun 2026 window it recorded a 70.6% win rate and a 5.9 annualized Sharpe on daily dollar P&L.
Research
My senior thesis, advised by Dr. David van Dijk, built an evaluation framework for whether language models can predict human survey responses, then used an ablation to separate behavior-learning from format-learning. I've also written up a comparison of robust optimization methods under adversarial data corruption, and earlier did condensed-matter physics research, developing an algorithm to extract superconducting gaps from photoemission spectra (2nd-place poster at a Berkeley Advanced Light Source meeting).
Earlier
At Axis (Oct 2025 – Jul 2026) I built most of the application layer myself: data ingestion, a provenance-tracked knowledge graph, and a forecasting backend.
Before Axis I was a junior quant strategist at Akuna Capital (Aug 2025 – Oct 2025) and a quant trading intern at Virtu Financial (summer 2023). In 2022, I built and shipped Convo, a mobile social app that reached 500+ users.
Before college I was a five-time AIME qualifier, reached USACO Gold, and was a USA Physics Olympiad Silver medalist.