Numfor Mbiziwo-tiapo is a research engineer with nine years' experience building and integrating machine learning systems across Windows and open-source runtimes. He shipped features on Microsoft's Windows Machine Learning team and contributed backend enhancements to ONNX Runtime, including experimental model APIs and NPU selection, showing comfort across core inference stacks. Earlier work at Google included designing and implementing IndexedDB putAll for Chrome and improving Linux perf coverage at scale, reflecting a strong testing and performance mindset. Now at the University of Washington's WEIRD Lab he focuses on AI and robotics research, bridging systems-level ML engineering with applied research. Colleagues would note his knack for turning integration complexity into actionable samples and tooling that make ML capabilities accessible to application developers.
9 years of coding experience
3 years of employment as a software developer
Bachelor's degree, Computer Science, Bachelor's degree, Computer Science at Rensselaer Polytechnic Institute
Contributions:134 reviews, 395 commits, 49 PRs in 1 year 6 months
Contributions summary:Numfor's contributions primarily involve integrating NuGet packages for Windows Machine Learning. They added and utilized NuGet packages in both C++ and C# code, modifying header files, sample helpers, and main application files to incorporate the packages. The user's work included adding file selection and inference features to a sample gallery application. The changes suggest the user was focused on integrating and utilizing Windows Machine Learning libraries within a broader application context, suggesting a focus on the front-end, and back-end integration.
ONNX Runtime: cross-platform, high performance ML inferencing and training accelerator
Role in this project:
Back-end Developer
Contributions:20 reviews, 12 commits, 55 PRs in 1 year 2 months
Contributions summary:Numfor's commits primarily involve modifying and extending the functionality of the ONNX Runtime. They added an experimental API to set model names, requiring changes to the core codebase and test files. Additionally, the user contributed to registering operators and addressed build and prefast errors within the project, contributing to the overall stability and functionality. The user also added the ability to select NPU devices.
runtimetrainingtensorflowai-frameworkaccelerator
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.
Request Free Trial
Numfor Mbiziwo-tiapo - Research Engineer at University of Washington