Takuya Shibayama
Machine Learning Application Engineer
Switzerland
Summary
🤩
Rockstar🎓
Top SchoolTakuya Shibayama is a machine learning application engineer with 13 years of experience who bridges high-performance computing from his PhD in plasma and solar physics to practical AI solutions. He developed GPU-scale simulations on systems like the K computer and Earth Simulator and now applies that HPC expertise to scale and optimize ML workflows. At Preferred Networks he contributed to notable open-source projects—implementing a pix2pix example in Chainer and improving the front end of the popular PaintsChainer colorization app—demonstrating both research-grade modeling and user-facing product work. Now based in Switzerland and working at Neural Concept, he combines deep numerical simulation knowledge with production ML engineering, and his TOEIC 890 highlights strong international collaboration ability.
13 years of coding experience
6 years of employment as a software developer
学士 天文学・天体物理学, 学士 天文学・天体物理学 at Kyoto University博士 天文学・天体物理学, 博士 天文学・天体物理学 at Nagoya University
Github Skills (16)
html10
neural-network10
machine-learning10
javascript10
user-interface10
deep-learning10
front-end-development10
python10
chainer10
computer-vision8
gpu7
bootstrap7
cupy7
cuda7
bootstra7
Programming languages (5)
Github contributions (5)
pfnet/PaintsChainer
Jan 2017 - May 2017
line drawing colorization using chainer
Role in this project:
Front-end Developer
Front-end DeveloperContributions:103 commits, 60 PRs, 111 pushes in 3 months
Contributions summary:Takuya primarily focused on enhancing the user interface of the PaintsChainer application. Their contributions involved adding a Twitter widget to the `index.html` file, indicating an effort to integrate social media functionality. Further, they modified the JavaScript file (`paints_chainer.js`) to address potential upload issues and improve the image processing workflow, suggesting a focus on refining the user's interaction with the core application features.
pythonchainercolorizationdrawingline-drawing
chainer/chainer
Apr 2017 - May 2018
A flexible framework of neural networks for deep learning
Role in this project:ML Engineer
ML EngineerContributions:17 commits, 3 PRs, 4 comments in 1 year 1 month
Contributions summary:Takuya implemented a pix2pix example within the Chainer deep learning framework. This involved creating and integrating components for image-to-image translation, as demonstrated by the addition of `train_facade.py`, `net.py`, and `updater.py` files. The changes include defining and utilizing Encoder, Decoder, and Discriminator networks along with corresponding training logic for the pix2pix model to process facade images.
cudapythonmxnetcaffe2flexible-framework
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.