Why markets, why models

Hey, I am Ansh. I build and test market models, and the thing I care about most is not fooling myself while doing it.

I did my undergrad at Arizona State in the Barrett Honors College, a double major in Computer Science and Mathematics, and graduated in May 2026 with a 4.0 and the Moeur Award. This fall I start the MSc in Financial Mathematics at the University of Chicago, on the trading track. The path from a CS and math kid who liked proofs to someone pointed at quant research ran through a simple realization: markets are a clean test of whether your model actually knows anything, because the feedback does not care how elegant your derivation was.

Most of what I work on sits at the intersection of Bayesian inference, stochastic processes, and careful evaluation. I am drawn to Bayesian methods because they force you to be explicit about what you believe and how much, and to regime switching and filtering problems because markets do not sit still and neither should your estimate of them. My honors thesis was about exactly this failure mode: how a Bayesian agent that has learned one regime keeps betting as if it still holds after the world changes, and what you can do to detect the shift faster and cap the damage.

The skepticism is the part I have had to learn the hard way. Early versions of my projects had numbers that looked great and did not survive contact with an out of sample test. Catching my own inflated results, understanding why they were wrong, and reporting the modest true number instead has taught me more than any clean success would have. A strategy that loses money for a reason I understand is worth more to me than one that wins for a reason I do not.

Outside of research, poker is a serious mathematical hobby I study with ranges, position, and disciplined sizing in mind. I train most days, I read a lot, and I am from Mumbai. I expect Chicago to be a demanding and useful chapter before I eventually point a lot of that learning back home.