Research that earns its latency.
We treat markets as noisy physical systems: observable, unstable, adversarial. A strategy is accepted only when it survives statistical pressure, implementation cost, and live-market constraints.
Discovery. Find the signal.
Signals begin as mathematical claims about microstructure, flow, and probability.
Falsification. Break it.
Ideas face null models, out-of-sample regimes, slippage, queue position, and capacity.
Control. Hold the risk.
Execution, sizing, inventory, and drawdown are modeled and run as one system.
A research lab with a market clock. Research and production share one source of time: no translation loss.
Three markets.
One discipline.
Novel strategies on shared infrastructure, from microsecond execution to systematic macro.
For people who prefer
hard problems.
Applied mathematics
for live markets.
Develop statistical models, test trading hypotheses, and turn first-class research into robust algorithms.
- Advanced work in mathematics, physics, statistics, computer science, or adjacent fields.
- Exceptional research record and strong programming taste.
- Comfort with ambiguity, noise, and empirical failure.
Systems for
scientific trading.
Build the software environment where data, models, simulation, execution, and accounting stay precise.
- Strong analytical programming in Linux or Unix environments.
- Experience with performance-sensitive systems and research tooling.
- Ability to work closely with scientists on hard infrastructure.