Zhenhui X
Algorithm Expert at Meituan
Haidian District, Beijing, China
Summary
🤩
Rockstar🎓
Top SchoolZhenhui X is an algorithm expert with 10 years of experience applying machine learning and data science to recommendations, advertising, and NLP across Meituan, Tencent, and Microsoft Research Asia. He blends academic rigor from Peking University and research stints at CUHK and UCLA with hands-on product impact, shipping LightGBM-based recommendation solutions and contributing practical notebooks to a popular recommenders best-practices repo. Known for bridging pretrained language models and tree-ensemble approaches, he repeatedly optimizes model pipelines from data preprocessing to evaluation for production-scale systems. Based in Haidian, Beijing, he brings a research-first mindset to industry problems, often surfacing simple, high-impact improvements that boost online recommendation performance.
10 years of coding experience
3 years of employment as a software developer
Master's degree, Data Science, Master's degree, Data Science at Peking UniversityBachelor's degree, Tang Aoqing Honors Program in Computer Science, Bachelor's degree, Tang Aoqing Honors Program in Computer Science at Jilin UniversityVisiting Scholar, Data Science, Visiting Scholar, Data Science at University of California, Los AngelesThe Chinese University of Hong Kong (CUHK)
Github Skills (9)
machine-learning10
jupyter-notebook10
recommendation-system10
python10
data-science10
lightgbm10
data-preprocessing9
auc8
binary-classification8
Programming languages (5)
Github contributions (5)
recommenders-team/recommenders
Mar 2019 - Mar 2019
Best Practices on Recommendation Systems
Role in this project:
Data Scientist
Data ScientistContributions:38 commits, 10 PRs, 18 pushes in 18 days
Contributions summary:Zhenhui's contributions primarily involve the initial implementation and refinement of LightGBM models within the recommendation system context. Their work focuses on a quick-start guide, demonstrating how to train LightGBM models using a dataset from the Criteo dataset for a binary classification task. The commits also include preprocessing of data, parameter setting and evaluating model performance to achieve high accuracy. Additional commits correct writing bugs and refine the display in the notebook.
recommendation-systemspythonjupyter-notebookoperationalizationmicrosoft
motefly/fairseq
Sep 2019 - Jan 2020
Facebook AI Research Sequence-to-Sequence Toolkit written in Python.
Contributions:143 commits, 10 PRs, 161 pushes in 4 months
nlpsequencepythonmachine-learningfacebook
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.