About Me

, , ,

I'm currently a System Research Scientist at Together AI, where I develop and optimize high-performance kernels for AI workloads.

Previously, I was a Postdoctoral researcher at Pacific Northwest National Laboratory (PNNL), where I was mentored by Dr. Ang Li. My work there focused on GPU-based dynamic system simulation, distributed quantum simulation on AMD GPUs, and efficient GPU acceleration for SPH algorithms.

I obtained my Ph.D. in Computer Science from the University of Utah, where I specialized in analyzing floating-point behavior in high-performance systems under the guidance of Prof. Ganesh Gopalakrishnan.

Previously, I earned my Master's in Computer Science at the University of Texas at Dallas under the guidance of Prof. Kyle Fox. My master's research focused on computational geometry, and my thesis was Approximating the Geometric Edit Distance.

I hold a Bachelor's degree from the Beijing University of Posts and Telecommunications in the Internet of Things.

My research focuses on numerical behavior and numerical correctness in high-performance accelerators, especially NVIDIA and AMD GPUs. I am particularly interested in floating-point error and exception analysis, binary instrumentation for HPC systems, tensor core numerical behavior, systematic reverse engineering of numerical phenomena, and performance improvement for AI workloads on GPUs.

My master's work also gave me a strong foundation in the rigorous analysis of algorithms and theoretical computer science, which continues to shape how I approach practical systems problems and intricate numerical algorithms.

Currently: System Research Scientist at Together AI