Jiacheng Yang
Research Assistant at University of Toronto
Old Toronto, Ontario, Canada
Summary
🤩
Rockstar🎓
Top SchoolJiacheng Yang is a PhD student and research assistant at the University of Toronto with nine years of experience focusing on efficient deep learning systems, especially deploying and accelerating DNNs on resource-constrained devices. His work spans algorithm design for training/inference, deep learning compilers, and practical system-level improvements, with internships at AWS and ByteDance and collaborations with MIT and Shanghai Jiao Tong University. He has contributed to open-source projects like MAgent by improving backend rendering and data handling for many-agent reinforcement learning visualization, reflecting a knack for bridging research and production systems. Jiacheng’s background includes published work on applied ML topics (e.g., learning analog circuit design and BERT for translation) and a consistent emphasis on making cutting-edge models practical for edge deployment.
9 years of coding experience
1 year of employment as a software developer
Master's degree, Computer Engineering, Master's degree, Computer Engineering at University of TorontoBachelor's degree, Computer Science, Senior, Bachelor's degree, Computer Science, Senior at Shanghai Jiao Tong University
Chinese, English, Japanese
Github Skills (9)
data-structures10
c-language10
cprogramming-language10
python10
reinforcement-learning10
data-structure10
deeplearning-ai9
deep-learning9
algorithms9
Programming languages (11)
Github contributions (5)
geek-ai/MAgent
Nov 2017 - Apr 2022
A Platform for Many-Agent Reinforcement Learning
Role in this project:
Back-end Developer
Back-end DeveloperContributions:37 commits, 4 PRs, 18 pushes in 4 years 5 months
Contributions summary:Jiacheng primarily focused on modifying the render backend to read combined data and included coordinate and obstacle data in the render process. These changes involved adjusting data structures like `Buffer`, `Config`, and `Frame`, and updating encoding/decoding methods within the `Text` and `Protocol` classes. Additionally, the user integrated changes to the server-side processing and rendering of map information. These modifications suggest a focus on improving the data handling and visualization aspects of the reinforcement learning platform.
reinforcement-learningmulti-agentdeep-learning
Kipsora/x86-64-Java-Miracle-Compiler
Mar 2017 - Jun 2017
Contributions:90 pushes, 5 branches in 3 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.