Yijie Bei is a versatile software engineer with 10 years of experience building large-scale ML systems, currently working on GenAI at Fireworks.ai after driving major ranking and ML infrastructure projects at Meta. He led a rewrite of Feed’s largest ranking model pipeline, achieving 5x training speedups and substantial inference gains, and built tooling for model understanding and verification that improved experimentation velocity and cost efficiency. A triple major from Duke in ECE, CS, and Math, Yijie combines deep research experience—publishing in CVPR and Nature Machine Intelligence—with hands-on production work across recommendation, training orchestration, and inference optimization. He moves fluidly between research and engineering, having implemented graph learning platforms, quantization/pruning toolkits, and large-scale data pipelines, and often tackles the hardest parts of model deployment rather than just the model itself. Notably, his background spans both algorithmic research (wave dark matter simulations, sampling libraries) and production ML at internet scale, reflecting a fast-learning, broad technical curiosity.
9 years of coding experience
7 years of employment as a software developer
Triple Major B.S.E. Electrical and Computer Engineering B.S. Computer Science B.A. Mathematics, Triple Major B.S.E. Electrical and Computer Engineering B.S. Computer Science B.A. Mathematics at Duke University
Cambridge-Singapore GCE-Alevel Certificate A-Level, Cambridge-Singapore GCE-Alevel Certificate A-Level at Victoria Junior College
Contributions:15 commits, 17 pushes, 3 branches in 4 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.