Mingkun Yang
PHD Candidate
Delft, South Holland, Netherlands
Summary
👤
Senior🎓
Top SchoolMingkun Yang is a Ph.D. candidate at TU Delft's Embedded and Networked Systems group with over a decade of experience bridging signal processing and machine learning. His research background spans Pedestrian Dead Reckoning and Visual-Inertial Odometry from UESTC, and he now focuses on human-centered approaches to interpret and improve AI model robustness. Co-supervised by experts across Delft and with strong foundations in communications engineering, he combines rigorous theory with practical system-building. Mingkun also works on computer vision problems such as scene text recognition and TextVQA, reflecting a versatile skill set across perception and multimodal reasoning. He aims to translate academic insights into tangible improvements in ML performance and reliability, often blending sensor fusion and human-in-the-loop methodologies. Based in Delft, he brings a global academic trajectory and a hands-on attitude toward making machine learning systems more trustworthy and resilient.
11 years of coding experience
Master of Engineering - MEng, Information and Communication, Master of Engineering - MEng, Information and Communication at University of Electronic Science and Technology of China
Github Skills (87)
aster10
python10
deep-learning10
gpu10
scene-text10
ocr10
machine-learning10
numpy9
neural-network9
computer-vision9
dialog9
tensorflow9
libtorch8
acceleration8
detectron8
Programming languages (8)
Github contributions (5)
ayumiymk/aster.pytorch
Jul 2019 - May 2020
ASTER in Pytorch
Contributions:1 release, 14 commits, 8 pushes in 9 months
pytorchaster
ayumiymk/DiG
Dec 2022 - Feb 2023
Contributions:9 commits, 3 pushes, 1 branch in 2 months
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.