Princeton Plasma Physics Laboratory

At Princeton Plasma Physics Laboratory (PPPL) — a U.S. Department of Energy national lab and one of the world's leading fusion research centers — I worked on advanced flow-visualization software for experimental physics. My focus was Background-Oriented Schlieren (BOS) imaging, a computer-vision technique that reconstructs density gradients in transparent media by measuring apparent distortion of a textured background.

I implemented BOS imaging algorithms in Python and MATLAB and applied PyTorch-based machine-learning models to the reconstruction pipeline, improving turbulence visualization accuracy by 35% over classical cross-correlation methods. To generate controlled experimental data, I built an embedded control system on Arduino with a backend logging infrastructure for artificial wavefront generation experiments, processing sequences of 10,000+ frames.

The research produced tangible IP and scholarship: I co-authored two peer-reviewed publications and filed a provisional patent for aerospace flow-visualization technology. This work directly informed my published research on real-time adaptive BOS imaging with spatiotemporal variability.