bio

Dataism Lab for Quantitative Finance

Where the execution research happens. Quantitative Researcher, Oct 2024 – present.

I've been a Quantitative Researcher at Virginia Tech's Dataism Lab for Quantitative Finance since October 2024, working under Dr. Ali Habibnia.

The through-line of my work here is Optimal Execution on real Bitcoin market microstructure data:

Built the benchmark execution algorithms, VWAP and TWAP, as the yardstick every learned strategy has to clear.
Engineered a Double Deep Q-Learning architecture in PyTorch for adaptive execution.
Built Mixture of Experts agents for execution, and the repeated-training evaluation protocol used to judge them.
Modelled execution efficiency statistically in Python (NumPy, Pandas, SciPy), with an emphasis on distributions rather than point estimates.
Built the verification suite that audited that pipeline end to end, validating the execution model numerically against an independent re-implementation, reconciling campaigns across devices, and decomposing which configuration choices were driving which result.

The lab is where I learned that the interesting part of quantitative research is rarely the model. It's the measurement. And, more often than I expected, it's going back to check whether the measurement was measuring what we said it was.