JetPrism: Generative AI Model for Particle Physics (In Progress)

• Pioneering generative AI models (diffusion and flow matching) to solve inverse problems in hadron spectroscopy, creating foundational models that correct detector distortions with unprecedented fidelity
• Building reusable ML infrastructure for the MLCommons open-source ecosystem, enabling FAIR-principled scientific discovery at scale

GPU-GEMF: Epidemiological Graph Modeling Platform (In Progress)

• Engineering GPU-accelerated HPC platform using approximation algorithms to optimize large-scale, network-based epidemiological simulations
• Delivering 10x+ runtime improvements, enabling complex population modeling and disease-spread forecasting for public health decision-making

PyTorchFire & FireStrategist: Wildfire Simulation & Mitigation

• Architected first GPU-accelerated, differentiable wildfire simulator achieving millisecond-level efficiency—orders of magnitude faster than existing methods; currently extending to wildland-urban interface (WUI) scenarios
• Developing firebreak placement algorithm integrating diffusion and RL models trained in real-time, creating a decision-making tool that generalizes across diverse climates

DT-SegNet: Deep Learning for Materials Science

• Built end-to-end deep learning pipeline combining YOLOv5 and SegFormer for automated microstructural analysis
• Achieved state-of-the-art performance, outperforming existing tools and creating high-throughput capabilities for materials discovery