Yizhou Sheng is a software engineer with eight years of experience building high-performance C++ systems, currently at Databricks and entering a CMU MS in Computer Science. He blends systems and research experience from USC’s Robotic Embedded Systems Lab—where he sped up simulation backends and compared autodiff tools—with production internships on Microsoft’s Direct3D team and NVIDIA. His open-source contributions include implementing SDF-to-mesh support in a popular differentiable physics simulator and adding DirectX 12 feature-support plumbing to Microsoft’s official headers. Comfortable across C++, CUDA, Python and cloud/distributed systems, he brings both low-level graphics and simulation know-how and an academic rigor reflected in near-perfect GPAs. Colleagues value his ability to translate research ideas into robust, tested code and to optimize architecture for both correctness and performance.
7 years of coding experience
3 years of employment as a software developer
Bachelor of Science - BS, Computer Engineering and Computer Science, GPA: 3.97, Bachelor of Science - BS, Computer Engineering and Computer Science, GPA: 3.97 at University of Southern California
Computer Information Systems, Computer Information Systems at De Anza College
Master of Science - MS, Computer Science, QPA: 4.14/4.00, Master of Science - MS, Computer Science, QPA: 4.14/4.00 at Carnegie Mellon University
Official DirectX headers available under an open source license
Role in this project:
Backend Developer
Contributions:1 review, 21 commits, 1 PR in 1 month
Contributions summary:Yizhou's primary contribution involves the development and implementation of a feature support class, `CD3DX12FeatureSupport`, within the DirectX headers. This class is designed to handle feature support checks for Direct3D 12, which includes various features like shader models, root signatures, and other options. The commits demonstrate the addition of the class, implementation of feature checks, and integration with the project's build system through modification of CMakeLists.txt. This work is crucial for ensuring compatibility and leveraging specific DirectX 12 capabilities.
Tiny Differentiable Simulator is a header-only C++ and CUDA physics library for reinforcement learning and robotics with zero dependencies.
Role in this project:
Back-end Developer
Contributions:10 commits, 1 PR, 5 comments in 2 months
Contributions summary:Yizhou contributed to the `tiny-differentiable-simulator` repository by adding and modifying core C++ code related to the Marching Cubes algorithm. These changes focused on implementing SDF to mesh conversion functionality. This involved incorporating new data structures, functions for linear interpolation, and macros for gradient calculations. The work allows for the generation of meshes from SDF representations within the simulator.
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.