Mengdi Wang (王梦迪)

PhD Candidate
School of IC, Georgia Institute of Technology
3142 Klaus Building
Email: mengdi.wang@gatech.edu
Resume | Google Scholar | LinkedIn | GitHub

About Me

I am a Ph.D. candidate in Computer Science at Georgia Tech, advised by Prof. Bo Zhu. I received my Bachelor degree of Computer Science at Peking University in 2020, advised by Prof. Baoquan Chen. My research focuses on building scalable, GPU-accelerated systems for multiscale physical simulation.

My work spans the full simulation pipeline, from physical representation and interface modeling to transport algorithms and large-scale solvers. I develop methods that bridge particle and grid formulations, enable low-dissipation transport through hybrid schemes, and achieve scalable performance via adaptive, matrix-free multigrid solvers on GPUs.

I aim to design unified simulation systems that achieve both physical fidelity and computational scalability, and to connect physics-based modeling with emerging AI-driven approaches for understanding and controlling complex physical processes.

Contributed open-source projects:

code code code code code

Publications

Cirrus: Adaptive Hybrid Particle-Grid Flow Maps on GPU
webpage paper code
ACM Transactions on Graphics (Proceedings of SIGGRAPH 2025)
Mengdi Wang, Fan Feng, Junlin Li, Bo Zhu



Real-Time Knit Deformation and Rendering
webpage paper
ACM Transactions on Graphics (Proceedings of SIGGRAPH 2025)
Tao Huang*, Haoyang Shi*, Mengdi Wang*, Yuxing Qiu, Yin Yang, Kui Wu



An interface tracking method with triangle edge cuts
webpage paper
Mengdi Wang, Matthew Cong, Bo Zhu

Journal of Computational Physics (Volume 520, 1 January 2025, 113504)



Hydrophobic and Hydrophilic Solid-Fluid Interaction
paper
Jinyuan Liu, Mengdi Wang, Fan Feng, Annie Tang, Qiqin Le, Bo Zhu
ACM Transactions on Graphics (Proceedings of SIGGRAPH Asia 2022)



A Clebsch method for free-surface vortical flow simulation
webpage paper
ACM Transactions on Graphics (Proceedings of SIGGRAPH 2022)
Shiying Xiong, Zhecheng Wang, Mengdi Wang, Bo Zhu



A Moving Eulerian-Lagrangian Particle Method for Thin Film and Foam Simulation
webpage paper code
ACM Transactions on Graphics (Proceedings of SIGGRAPH 2022)
Yitong Deng, Mengdi Wang, Xiangxin Kong, Shiying Xiong, Zangyueyang Xian, Bo Zhu



Thin-Film Smoothed Particle Hydrodynamics Fluid
webpage paper code
ACM Transactions on Graphics (Proceedings of SIGGRAPH 2021)
Mengdi Wang, Yitong Deng, Xiangxin Kong, Aditya H. Prasad, Shiying Xiong, Bo Zhu



Visual Data Analysis and Simulation Prediction for COVID-19
paper
Published in arXiv preprint, 2020
Baoquan Chen, Mingyi Shi, Xingyu Ni, Liangwang Ruan, Hongda Jiang, Heyuan Yao, Mengdi Wang, Zhenhua Song, Qiang Zhou, Tong Ge.

Experience

Game Engine Research Intern
TikTok Inc.
San Jose, CA
2025.5 ~ Present


Research Intern
Lightspeed Studio, Tencent America
Los Angeles, CA
2024.5 ~ 2024.8


Research Intern
NVIDIA Research
Santa Clara, CA (remote)
2022.6 ~ 2022.9