Feng Tian
Senior Software Engineer at Intel Corporation
Minhang District, Shanghai, China
Summary
🤩
RockstarFeng Tian is a Senior Software Engineer with 14 years of experience, currently at Intel in Shanghai, combining low-level systems engineering with advanced model optimization for deep learning. He contributes to high-profile open-source projects—improving EDK II firmware drivers for USB, SCSI and storage (UFS/eMMC) while also enhancing Intel's Neural Compressor for state-of-the-art low-bit quantization and pruning across TensorFlow, PyTorch, and ONNX. His work spans bug fixes to performance optimizations and robust save/restore and tuning strategies, reflecting a pragmatic focus on production reliability and reproducible ML workflows. Comfortable across firmware, driver stacks and ML toolchains, he bridges hardware-aware software considerations with algorithmic compression techniques. Colleagues rely on him for hard-to-debug platform issues and for making cutting-edge quantization methods usable in real deployments.
14 years of coding experience
Github Skills (18)
post-training10
uefi10
pytorch10
python10
machine-learning10
usb10
edk210
c1110
scsi10
c1710
driver10
tensorflow10
device-driver10
model-optimization10
quantization10
Programming languages (8)
Github contributions (5)
intel/neural-compressor
Jun 2020 - 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:
Back-end Developer & ML Engineer
Back-end Developer & ML EngineerContributions:21 releases, 209 reviews, 234 commits in 2 years 7 months
Contributions summary:Feng contributed to bug fixes and enhancements within the codebase, specifically focused on improving the save/restore functionality for MSE and Bayesian tuning strategies. They addressed issues related to the inspect_tensor API for the TensorFlow adaptor. Furthermore, the user made several example code enhancements and fixes, primarily addressing issues with examples within the PyTorch environment. Their work touches on the core aspects of the project, including model compression techniques and quantization strategies, highlighting their understanding of the low-bit LLM quantization methods in the project.
knowledge-distillationauto-tuningcompressorsparsityintel
tianocore/edk2
Nov 2011 - May 2017
EDK II
Role in this project:Back-end Developer
Back-end DeveloperContributions:172 commits, 61 pushes, 1 branch in 5 years 6 months
Contributions summary:Feng primarily contributed to the EDK II project, focusing on improvements to the ATA/ATAPI, USB, and SCSI bus drivers. Their work included fixing bugs in memory handling, optimizing performance in the XHCI driver, and addressing issues related to command execution and device compatibility. They also added support for UFS and EMMC devices and enhanced the error handling and compliance with the UEFI specifications.
uefipythonfirmwarec
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.