Program · FIN·01
We build systems that read markets as they move, learning the latent structure beneath price, liquidity, and risk, and acting on it in real time. The aim is not a single strategy but a model of market dynamics general enough to transfer across assets and regimes.
What we work on
A program is a small set of hard, coupled problems. These are the ones this group is working on now.
Most quantitative systems predict a number. We train systems to represent the process that produces the number: the flow of capital, the propagation of information, the feedback between participants. A model of mechanism degrades gracefully when the world changes, because it has learned structure rather than correlation.
Equities, rates, credit, and commodities are not separate problems. They are views of one coupled system. A shared representation lets a signal learned in one market inform decisions in another, and lets risk be measured where it actually lives, across the whole portfolio rather than inside a single book.
Every position is an experiment. Fills, slippage, and realised risk feed back into the model within the trading day, tightening the loop between what the system believes and what the market confirms.
Signals from the program
Selected writing from this program
Collaborate
We hire researchers and engineers who want to push one of these programs forward, and partners who want to put the results to work.
