Selected Work Experience
Developing AI systems for on-site waste detection and sorting. Working on multi-object tracking and model robustness against domain changes & class flickering.
Applied AI Research Engineer
Developed Squeeze-and-Excitation + feature genuinization model for fake audio detection for Bixby. Evaluated on ASVSpoof 2019, outperforming other models using Mel Spectrogram features.
Supervisor: Sourabh Tiwari
ML Research Engineer
Developed a method to control text-to-image generation system using diffusion models & guided adversarial prompts for effective data generation. Achieved data-efficient generations, outperforming others in low-data regimes across various tasks & distribution shifts.
Supervisor: Prof. Amir Zamir & Dr. Teresa Yeo
Research Fellow & Master's Thesis
Graduate Research Assistant
University of Texas at Austin
RBT350 Gateway to Robotics: created & graded homework and exams, led hands-on review sessions, taught robot assembly (building the Stanford Pupper Robot from scratch) and control (PD control, forward and inverse kinematics), held office hours, and managed the course webpage.
Supervisor: Dr. Roberto Martin-Martin (Fall 2023), Dr. Mitchell Pryor (Spring 2024)
Graduate Teaching Assistant
MacEwan University
Developed mathematical equation recognition model for first-year calculus assignments and exams using YOLOR.
Supervisor: Prof. Jeffrey Davis
MITACS Globalink Intern
