Experience

Currently pursuing my PhD at the Wyant College of Optical Sciences, I work on cutting-edge applications of machine learning to real-world problems in industrial and astronomical optics. My research spans from autonomous mirror alignment for large space telescopes to developing physics-informed neural networks for advanced optical systems. It's fascinating work that bridges the gap between computational intelligence and physical optics.

I spent summer 2026 at Brookhaven through the Graduate Research Internship Program, in the NSLS-II Optical Metrology Group with Tianyi Wang as my mentor. Wang and Mourad Idir wrote the dwell-time algorithm that RUDO builds on, and this was my chance to run RUDO on their ion beam figuring system instead of in simulation. It planned every correction of a campaign on a 140 mm X-ray mirror, taking the measured figure error from 669 nm to 8.8 nm RMS.

At this innovative startup, I applied advanced ML techniques to analyze fluorescence data for miniaturized spectrometer products. I led a team of 6 student engineers as Project Sponsor, achieving a 90% improvement in manufacturing time for the F1 Drug Detector. The experience taught me how to balance research-driven development with practical engineering constraints in a fast-paced startup environment.

Working under Dr. Chi-Kwan Chan, I contributed to cutting-edge astronomy research on black holes. I developed a production-grade cloud computing environment using Docker and Kubernetes to support the team's computational needs, and created a preliminary neural network auto-encoder to assist in removing noise from black hole images. This was my first deep dive into interdisciplinary high-performance computing and astronomical research.

As Scrum Master, I led a team of System Engineers and Computer Scientists to build an in-house Simulation Engine and Architecture Creation Tool from the ground up. This role taught me valuable lessons in agile project management, cross-functional collaboration, and the importance of making informed technical decisions through thorough trade studies.

Teaching a class of 160-180 students gave me a deep appreciation for clear communication and well-designed code. I emphasized clean code practices, modular design patterns, and comprehensive testing. Working with students one-on-one during office hours and designing challenging assignments that required 100% branch coverage helped me develop my ability to break down complex concepts into digestible pieces.

For four years, I tutored students in mathematics ranging from Algebra I to Calculus I. After earning certification through the University of Arizona, I maintained a private practice that helped students improve their exam scores by 10-30%. This experience reinforced my belief that anyone can learn complex subjects with the right guidance and practice.