Will Usher is a Scientific Visualization Engineer with 13 years of experience building high-performance CPU and GPU ray tracing systems for scientific visualization, film, and web-based rendering. He led GPU and MPI-parallel rendering work in OSPRay at Intel and now applies Physics AI to accelerate analysis and visualization at Luminary. An active open-source contributor and core developer of OSPRay, he created ChameleonRT and tools like ispc-rs and WebGPU/WebGL visualization projects that the community uses for terascale data. His research includes 15 journal papers (9 in IEEE TVCG), a patent, and pioneering work on terascale WebGPU visualization, reflected in a Google Scholar h-index of 19. He combines deep systems expertise—porting ISPC to macOS/Arm64 and improving pbrt and GLM build systems—with practical skills in distributed MPI buffering and cross-team production deployment. Based in Los Angeles, he holds a Ph.D. in Computer Science and pairs rigorous research with production-grade engineering.
13 years of coding experience
7 years of employment as a software developer
The University of Utah
Bachelor's Degree, Physics, Bachelor's Degree, Physics at University of California, Riverside
An Open, Scalable, Portable, Ray Tracing Based Rendering Engine for High-Fidelity Visualization
Role in this project:
Back-end Developer
Contributions:962 commits, 7 PRs, 75 comments in 7 years 8 months
Contributions summary:Will made contributions related to safe buffered MPI comm manager. This involved code changes within the ospray/mpi/MPICommon.cpp, ospray/mpi/MPICommon.h and ospray/mpi/MPIDevice.cpp files. Their work included modifications to the MPI interface, setting up of buffers, sending and receiving of messages.
Contributions:1 review, 16 commits, 2 PRs in 6 years 5 months
Contributions summary:Will primarily focused on improving the Intel® Implicit SPMD Program Compiler (ISPC) by enhancing its compatibility, functionality, and build processes. Their contributions include enabling support for macOS and Arm64 targets, which involved modifying the target triple configuration. They also addressed target name collisions, optimized dependency handling within the CMake build system, and integrated definitions for GPU targets. Furthermore, the user implemented a feature for setting local group sizes for kernels and updated the build system to handle oneAPI installations.
cpusimdavximplicitavx512
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.