Yuan Zhang
Process Engineer at Intel Corporation
Hillsboro, Oregon, United States
Summary
🤩
Rockstar🎓
Top SchoolYuan Zhang is a process engineer and PhD-trained mechanical engineer with 9 years of multidisciplinary experience spanning materials science, thin film deposition (PVD/CVD), characterization, finite element simulation, and data analysis. Currently at Intel, he focuses on deposition processes and brings a strong research pedigree with projects funded by Microsoft, DOE, ONR and ARPA‑E. He also contributes to open-source ML and HPC tooling—improving SYCL backend Windows support for the widely used llama.cpp inference project and adding quantization examples to Intel Neural Compressor—demonstrating a rare mix of device-level process expertise and practical ML/DevOps skills. Comfortable moving between lab, simulation, and code, he leverages experimental rigor and reproducible workflows to accelerate materials and ML-enabled process development.
9 years of coding experience
7 years of employment as a software developer
Doctor of Philosophy - PhD, Mechanical Engineering, Doctor of Philosophy - PhD, Mechanical Engineering at University of Houston
English, Chinese
Github Skills (42)
pytorch10
quants10
gnu-make10
github-ci10
performance-monitor10
c-language10
performance-analytics10
python10
jupyter10
windows10
llama10
gpu-programming10
machine-learning10
build-system10
inference10
Programming languages (8)
Github contributions (5)
oneapi-src/oneAPI-samples
Oct 2020 - Jun 2022
Samples for Intel® oneAPI Toolkits
Role in this project:
ML Engineer
ML EngineerContributions:31 reviews, 9 commits, 23 PRs in 1 year 8 months
Contributions summary:Yuan primarily contributed to the development and maintenance of samples related to Intel's Low Precision Optimization Tool (LPOT) and Intel Neural Compressor within the context of Tensorflow. Their work involved creating, modifying, and integrating sample code, including the creation of notebooks demonstrating model quantization techniques. The user also addressed issues by renaming components and updating code to align with the changing names of the LPOT tool. This involved modifications to various files including notebooks, python scripts and configuration files.
numbatoolkitsswrepoaplpower-management
intel/neural-compressor
Mar 2022 - Jan 2023
SOTA low-bit LLM quantization (INT8/FP8/INT4/FP4/NF4) & sparsity; leading model compression techniques on TensorFlow, PyTorch, and ONNX Runtime
Role in this project:ML Engineer
ML EngineerContributions:19 reviews, 18 commits, 23 PRs in 10 months
Contributions summary:Yuan's commits focus on adding examples demonstrating the use of the Intel Neural Compressor (INC) within the context of model quantization and performance comparison. These examples cover different frameworks like TensorFlow and PyTorch, specifically showcasing the application of INC on AlexNet and ResNet50 models. The contributions include Jupyter notebooks detailing model quantization and performance analysis, and scripts comparing FP32 and INT8 model performance.
knowledge-distillationauto-tuningcompressorsparsityintel
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.