Yunfei Ma
Senior Staff Engineer, MEMS Design at Syntiant Corp.
Greater Chicago Area United States
Summary
🤩
Rockstar🎓
Top SchoolYunfei Ma is a Senior Staff MEMS Design Engineer with over a decade of experience turning multi-physics FEA and DFSS theory into high-volume MEMS microphone products and novel MEMS mirror technologies. Holding a Ph.D. in Mechanical Engineering from the University of Washington, he has led R&D from concept through mass production at Knowles and now drives MEMS design at Syntiant. His background spans advanced CAE and structural analysis at GE and consultancy for Microsoft Surface, grounding his work in rigorous simulation and manufacturability. Yunfei also contributes to open-source AI performance projects, optimizing deep learning models and frameworks for high-efficiency inference on Intel platforms and LF AI & Data initiatives. He combines deep academic training with hands-on optimization experience, often finding cross-disciplinary solutions at the intersection of mechanical design and high-performance computing. Colleagues rely on him for practical DFSS implementations that reduce cycle time and improve yield in complex MEMS products.
6 years of coding experience
17 years of employment as a software developer
Ph.D. Mechanical Engineering, Ph.D. Mechanical Engineering at University of WashingtonM. Eng. Engineering Mechanics, M. Eng. Engineering Mechanics at Huazhong University of Science and Technology
English, Chinese
Github Skills (24)
python10
machine-learning10
inference10
deeplearning-ai10
deep-learning10
tensorflow10
performance-optimization10
object-detection10
computer-vision10
algorithm9
code-optimization9
algorithms9
cpu9
optimisation9
ai9
Programming languages (4)
Github contributions (5)
DeepRec-AI/DeepRec
Jul 2020 - Sep 2020
DeepRec is a high-performance recommendation deep learning framework based on TensorFlow. It is hosted in incubation in LF AI & Data Foundation.
Role in this project:
ML Engineer
ML EngineerContributions:15 commits in 1 month
Contributions summary:Yunfei primarily focused on fixing bugs and improving the performance of the deep learning framework. They addressed issues related to bfloat16 support in unit tests and softmax operations. The user also implemented optimizations within the non-max suppression algorithm to improve efficiency and addressed issues related to MKL (Math Kernel Library) integration and disablement. Furthermore, they ported dequantization-related changes.
pythondeep-learningrecommendationdistributed-trainingmachine-learning
intel/ai-reference-models
Mar 2020 - May 2020
Intel® AI Reference Models: contains Intel optimizations for running deep learning workloads on Intel® Xeon® Scalable processors and Intel® Data Center GPUs
Role in this project:ML Engineer
ML EngineerContributions:6 commits in 2 months
Contributions summary:Yunfei made several contributions focused on optimizing and improving object detection models within the Intel AI reference models repository. They added support for a 1200x1200 input size for SSD-ResNet34 inference, addressing performance and accuracy. They also fixed an issue causing the SSD-ResNet34 model to fail in nightly testing, and made code adjustments for the latest SSD-ResNet .pb file format. Furthermore, they introduced optimizations to enhance performance when utilizing a high number of cores on a single node.
optimizationsprocessorstensorflowzoomodel-zoo
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.