Yuanming Hu

Yuanming Hu

胡渊鸣 博士


Co-founder and CEO of Meshy (formerly known as Taichi Graphics, rebranded for a snappier, more memorable name).

Ph.D. in computer science (MIT CSAIL), advised by Frédo Durand and Bill Freeman.

I graduated with honor from Tsinghua University (Yao class) in July 2017. I worked on deep learning and computer vision, during my internship with Stephen Lin at Microsoft Research Asia. My undergrad thesis is on automatic (differentiable) photo post-processing using reinforcement learning & GANs (TOG & SIGGRAPH 2018). I completed my master thesis (The ChainQueen Differentiable Physical Simulator) with Wojciech Matusik in November 2018. My research has been partly supported by an Edwin Webster fellowship, a Snap Research fellowship, an Adobe Research fellowship, a Facebook Research fellowship, and a GPU gift from NVIDIA.

import taichi as ti

Update March 2021: I completed my Ph.D. thesis defense :-) After graduating from MIT, I serve as the co-founder and CEO of Meshy. At Meshy we continue the development of Taichi Lang, and have been building new products including Taitopia Render for product visualization and Meshy AI for 3D generative AI.

Update July 2022: The Taichi thesis is awarded MIT EECS 2022 George M. Sprowls Thesis Award and SIGGRAPH 2022 Outstanding Doctoral Dissertation Award Honorable Mention.

Taichi news:

  • ETH Zurich adopted Taichi to teach physically-based simulation in fall 2021! The vast majority of the students used the Taichi for their final projects. [Read the full story].
  • Made a video collection of recent Taichi community contributions.
  • Check out Taichi Voxel Challenge with 100+ submissions.
  • Taichi v1.0.3 is released! Try the latest version:
pip install --upgrade taichi
ti gallery

Email: yuanmhu at gmail dot com


  • PhD in EECS, MIT

  • MSc in EECS, MIT

  • BEng in Computer Science, Yao class, Tsinghua University


  • Computer graphics
  • Compilers
  • (Differentiable) Physical simulation
  • High-performance computing
  • Differentiable programming
  • Deep learning
  • Computational photography