Yonghyun Cho
Software Engineer at Amazon Leo
Issaquah, Washington, United States
Summary
🤩
Rockstar🎓
Top SchoolYonghyun Cho is a software engineer with 7 years of professional experience, currently building cloud services at Amazon Web Services after a master's in computer science from USC. He blends practical web and software development experience from Samsung SDS with research-focused machine learning work as a USC research assistant, giving him a strong bridge between production systems and ML experimentation. On open source, he contributed RL algorithm implementations and minibatch support to the well-regarded garage toolkit, demonstrating hands-on expertise in PyTorch, PPO, and policy modules. A multi-hackathon prize winner, he brings a competitive, product-oriented approach to prototyping and shipping ML features. Based in Los Angeles, he focuses on turning research-grade RL components into reliable, testable code that scales in real-world cloud environments.
7 years of coding experience
13 years of employment as a software developer
Master's degree Computer Science, Master's degree Computer Science at University of Southern CaliforniaBachelor of Science in engineering Computer Science, Bachelor of Science in engineering Computer Science at Handong Global University
Github Skills (10)
algorithm10
pytorch10
ppp10
reinforcement-learning10
ml9
machine-learning9
mle9
rep8
repr8
testing7
Programming languages (2)
Github contributions (5)
rlworkgroup/garage
Jul 2019 - May 2020
A toolkit for reproducible reinforcement learning research.
Role in this project:
ML Engineer
ML EngineerContributions:10 commits, 13 PRs, 240 pushes in 9 months
Contributions summary:Yonghyun primarily contributed to the development of reinforcement learning algorithms and modules within the PyTorch framework. Their contributions focused on implementing Gaussian MLP modules and policies, as well as integrating the VPG (Vanilla Policy Gradient) and PPO (Proximal Policy Optimization) algorithms, showcasing expertise in designing and implementing core RL components. These additions likely enhanced the toolkit's capabilities for training and evaluating RL agents, as evidenced by the addition of test code for benchmarking. The user further added minibatch support for the VPG and PPO implementations.
reinforcement-learningrl-algorithmsreproducibilitypytorchtensorflow
luiszugasti/octo-news-reel
Sep 2019 - Sep 2019
Contributions:37 pushes, 2 branches in 1 day
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.