I'm an applied mathematician and numerical methods engineer with a background in physics, applied mathematics, and scientific computing.
My academic training was in nonlinear dynamics — PhD in physics and MS in applied math at the University of Michigan. My thesis applied PDE-constrained optimal control to fluid mixing, with related work on active matter and acoustic droplet vaporization. Building numerical solvers from scratch — finite-difference, spectral, gradient-based PDE control — shaped how I approach engineering: start from the mechanism, make the assumptions explicit, and let the math discipline the design.
Out of graduate school I spent two years as a machine learning engineer at a fintech startup, building a suite of time-series forecasting models — ARIMA, state-space and Kalman filters, gradient boosting, LSTM/RNN sequence networks — and applying quadratic-programming optimization to portfolio asset allocation. The work translated the applied-math lens from PDEs into discrete dynamical systems and inference.
Since 2021 I've been at Quadric, a custom AI inference processor startup, as the in-house numerics lead. I've implemented floating-point hardware units in Verilog, built company-wide numerical testing methodology, developed fixed-point validation tooling and ONNX-to-silicon error tracing, and led the first Ax=b solvers on the chip.
