Liangchao Wu
Algorithm Engineer at ByteDance
Beijing, China
Summary
👤
Senior🎓
Top SchoolLiangchao Wu is a Senior Algorithm Engineer with 10 years of experience designing large-scale ranking, recommendation, and monetization systems that translate algorithmic advances into measurable business growth. He has led traffic and mixed-ranking strategy at TikTok Live and search ads, pioneering score-shading extensions to auction theory, unified UX value modeling, and uplift-based intervention strategies that balance short-term revenue with long-term retention. Previously he launched industry-scale federated learning in ads at ByteDance and built creator monetization and distribution systems at Xiaohongshu that drove double-digit revenue and efficiency gains. A prolific experimenter and team builder, he has shipped 50+ A/B tests, led 0→1 products, and contributed to the bytedance/fedlearner open-source framework. He combines deep causal and auction-theoretic thinking with pragmatic engineering and clear communication, and publishes extensively on ads tech and ranking at wulc.me.
10 years of coding experience
Master, Computational Science, Master, Computational Science at 华南理工大学Bachelor, Electronic Engineering, Bachelor, Electronic Engineering at South China University of Technology
Chinese, Chinese
Stackoverflow
1reputation
0reached
0answers
0questions
Github Skills (8)
machine-learning10
python10
modeling9
trainings9
web-framework9
deploying8
bash7
data-filtering6
Programming languages (3)
Github contributions (5)
bytedance/fedlearner
Aug 2020 - Aug 2020
A multi-party collaborative machine learning framework
Role in this project:
Back-end Developer
Back-end DeveloperContributions:5 commits, 7 PRs in 13 days
Contributions summary:Liangchao primarily focused on fixing bugs and adding features related to the `fedlearner` framework's core functionality. They addressed issues in the sparse estimator, including model meta export errors and feature column handling. Furthermore, they made changes to the trainer master and worker scripts by adding data filtering options and environment variables for model serving. Their work demonstrates involvement in the trainer's inner workings and deployment scripts.
pythonpartydata-sciencedeep-learningmachine-learning
WuLC/show-me-the-code
Apr 2016 - Oct 2016
Contributions:45 commits, 43 pushes, 1 branch in 5 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.