Michael Welter
Controls Technician at Amazon Web Services (AWS)
Tokyo, Japan
Summary
🤩
Rockstar🎓
Top SchoolMichael Welter is a Data Center Operations Server Engineer based in Tokyo with a decade of hands-on experience spanning HVAC maintenance, preventative maintenance planning at MIT, and current server engineering at AWS. He blends operational rigor with applied machine learning, having contributed a neural-network–based face alignment tracker to the open-source opentrack project that added recurrent models, uncertainty estimation, and eye-open/close classification. Comfortable in both physical infrastructure and ML workflows, he has delivered predictive models for building energy use through an Omdena AI-for-Good engagement. Michael holds a BS in Information Technology and brings practical systems thinking from facilities to cloud-scale data centers. Colleagues rely on him for durable, safety-conscious solutions that bridge hardware reliability and intelligent software.
10 years of coding experience
4 years of employment as a software developer
Bachelor of Science - BS, Information Technology, Bachelor of Science - BS, Information Technology at University of Phoenix
Github Skills (11)
neural-network10
computer-vision10
opencv10
machine-learning10
c-language10
cprogramming-language10
ai10
onnx9
cpp9
c179
qt7
Programming languages (6)
Github contributions (5)
opentrack/opentrack
Mar 2021 - Nov 2022
Head tracking software for MS Windows, Linux, and Apple OSX
Role in this project:
ML Engineer
ML EngineerContributions:15 reviews, 22 commits, 36 PRs in 1 year 7 months
Contributions summary:Michael implemented a face alignment-based tracker using a neural network. They integrated an AI-powered model, referencing external code for model generation. The user made multiple changes to the neural network tracker, fixing a potential crash related to tiny bounding boxes, displaying inference time and relevant elements, and incorporating a new coordinate system. They also added support for recurrent models, uncertainty estimation, and an eye-open/close classifier, refining the tracking logic and preview display.
linuxwindowsc-plus-plusflight-simulatorlinux-app
opentrack/neuralnet-tracker-traincode
Mar 2021 - Nov 2021
Training of machine learning models for the tracker
Contributions:1 release, 15 commits, 32 pushes in 7 months
machine-learning-models3d-face-alignmentdeep-learningface-alignmentfacealignment
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.