Ph.D. Research, University of Southern California, ISE, 2025
Proposed an online adaptive ADMM framework that updates penalty parameters via exact hypergradients, enabling stable and fast convergence for large-scale constrained machine learning and optimization problems. Led by Professor Meisam Razavaiyayn
Undergraduate Research, Cornell University, CIS (BURE), 2024
Built large-scale geospatial vision–language datasets by generating detailed captions from internet imagery using LLaVA-1.5 and LLaMA-3, enabling distributed training of multimodal models with over one million aligned image pairs. Led by Professor Kavita Bala and Professor Bharath Hariharan
Applied Optimization Project, Cornell University, 2023
Leader of the Prelim Scheduling subteam for the Cornell Scheduling Team using integer programming, incorporating fairness constraints and ML-assisted enrollment predictions. Led by Professor David Shmoys.
Independent Study, Cornell University, 2022
Designed an NVIDIA Omniverse extension for architectural visualization, integrating machine learning pipelines that translate blueprints and hand-drawn sketches into ML-generated 3D geometry within professional graphics workflows. Led by Professor Donald P. Greenberg.