Jingji Chen is a Research Scientist at ByteDance Seed specializing in large-scale LLM training systems and frameworks, with nine years of experience spanning HPC, distributed computing, ML systems, and graph systems. He holds a BE from Tsinghua and a PhD from Purdue, and has built high-performance ML and graph systems using OpenMP, MPI, and CUDA with C++ and Python. His work bridges rigorous academic research—published at top computer systems conferences—with production-oriented system design, gained through roles at Purdue, USC, and a Microsoft research internship. Based in Shanghai, he combines deep systems-level optimization skills with practical experience scaling training pipelines for modern large language models, and maintains an active publication record that highlights novel performance engineering insights.
9 years of coding experience
5 years of employment as a software developer
Doctor of Philosophy - PhD, Computer Engineering, 3.96, Doctor of Philosophy - PhD, Computer Engineering, 3.96 at University of Southern California
Doctor of Philosophy - PhD, Computer Science, Doctor of Philosophy - PhD, Computer Science at Purdue University
Bachelor of Engineering - BE, Computer Science, Bachelor of Engineering - BE, Computer Science at Tsinghua University
Code for our ACL 2019 long paper: "Ensuring Readability and Data-fidelity using Head-modifier Templates in Deep Type Description Generation".
Contributions:12 commits, 2 PRs, 8 pushes in 3 years 6 months
pytorchmodifierdeep-learningheadlong
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.