George Nash
AI Framework Engineer at Intel Corporation
Port Orchard, Washington, United States
Summary
🤩
Rockstar🎓
Top SchoolGeorge Nash is an AI Framework Engineer with 10+ years building high-performance ML inference and IoT software, currently driving ONNX Runtime kernel and execution-provider optimizations at Intel. He blends low-level C/C++ and SIMD expertise—shipping AVX10/AVX512-FP16 and AMX-backed kernels that outperform prior implementations—with a background in firmware, microcontrollers, and electrical test equipment. His open-source contributions include accelerating ONNX Runtime training samples and integrating CUDA/oneDNN providers, demonstrating practical experience bringing hardware acceleration to production ML stacks. Earlier roles at Qualcomm and in IoTivity/AllJoyn show deep protocol, bindings, and tooling experience across C, Java, JavaScript, and C#, plus a track record of improving test infrastructure and documentation. Colocated in Port Orchard, WA, he pairs systems-level performance tuning with developer-focused automation and long-running stewardship of complex open-source projects.
10 years of coding experience
12 years of employment as a software developer
University of Alaska Fairbanks
English, Russian
Github Skills (11)
neural-network10
machine-learning10
deeplearning-ai10
deep-learning10
onnx10
edn10
hardware-acceleration10
cprogramming-language9
c-language9
pytorch9
cuda8
Programming languages (6)
Github contributions (5)
microsoft/onnxruntime
Dec 2020 - Nov 2022
ONNX Runtime: cross-platform, high performance ML inferencing and training accelerator
Role in this project:
ML Engineer
ML EngineerContributions:68 reviews, 26 commits, 33 PRs in 1 year 10 months
Contributions summary:George focused on enhancing the ONNX Runtime's training capabilities, particularly for machine learning inferencing and training acceleration. Their work involved enabling and optimizing training samples, such as the MNIST example, by integrating support for different execution providers like CUDA and oneDNN, allowing for CPU and GPU acceleration. The commits demonstrate expertise in integrating hardware acceleration and optimizing performance for deep learning models. They also worked on preventing memory leaks and improving the efficiency of the training pipeline.
runtimetrainingtensorflowai-frameworkaccelerator
georgen117/onnxruntime
Dec 2020 - Sep 2023
ONNX Runtime: cross-platform, high performance ML inferencing and training accelerator
Contributions:36 pushes, 47 branches in 2 years 9 months
pytorchdeep-learningruntimemachine-learningonnx
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.