Yuefeng Zhang
Manager, Senior Data Scientist
Naperville, Illinois, United States
Summary
🤩
Rockstar🎓
Top SchoolYuefeng Zhang is a Manager and Senior Data Scientist with nine years of focused experience in machine learning, deep learning, computer vision, and NLP, currently leading R&D for multi-camera vision systems at McDonald’s. He designs and delivers production-grade solutions for object detection, tracking, re-identification and pose estimation to automate food-order verification, drive-thru workflows, and individualized work-utilization analytics. Prior roles at Capital One, Wavicle, and SMS Assist reflect a strong background in large-scale real-time data pipelines, probabilistic matching, pricing prediction, and end-to-end ML lifecycle delivery on cloud platforms like GCP, AWS, Databricks and SageMaker. An active open-source ML engineer, he contributed key functionality to the popular deepctr-torch repository (adding DIN and advanced activation support), underscoring his practical expertise in CTR modeling. With a PhD in Computer Science and two decades of engineering leadership stretching back to Motorola, he blends academic rigor with hands-on systems architecture and cross-functional team leadership. He’s driven by pragmatic innovation—“build the kernel of the world”—turning research ideas into robust, deployable AI systems.
9 years of coding experience
25 years of employment as a software developer
Ph.D., Computer Science, Ph.D., Computer Science at Western University
Github Skills (9)
model-building10
pytorch10
activation-function10
deep-learning10
python10
modeling10
model-driven10
model-driven-development10
machine-learning9
Programming languages (11)
Github contributions (5)
shenweichen/DeepCTR-Torch
Dec 2019 - Mar 2020
【PyTorch】Easy-to-use,Modular and Extendible package of deep-learning based CTR models.
Role in this project:
ML Engineer
ML EngineerContributions:8 commits, 12 PRs, 2 pushes in 3 months
Contributions summary:Yuefeng made significant contributions to the `deepctr-torch` repository, primarily focused on enhancing and expanding the model's functionality. Their work included fixing accuracy issues in log loss metrics, adding and refactoring activation layers like Dice, and supporting multiple activation functions. Furthermore, the user integrated the Deep Interest Network (DIN) model, showcasing their expertise in implementing advanced deep learning architectures for CTR prediction.
pytorchloss-functionsdeep-learningfibinetcomputer-vision
ZhangYuef/Hit-the-Mole
Aug 2016 - Sep 2024
Contributions:2 pushes, 1 comment, 1 issue in 8 years 3 months
gamewindowswindows-gamebupthit
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.