Daochen Zha
Senior Machine Learning Engineer at Airbnb
Houston, Texas, United States
Summary
🤩
Rockstar🎓
Top SchoolDaochen Zha is a Senior Machine Learning Engineer with nine years of experience bridging research and production ML, currently leading ML efforts at Airbnb and completing a PhD at Rice University. His background spans reinforcement learning, anomaly detection, AutoML and graph representation learning, with internships at Meta and Kuaishou that informed both large-scale systems and game-playing RL research. He has contributed to high-impact open-source projects—helping refine DouZero (ICML 2021 DouDizhu RL) and adding core brute-force search capabilities to TODS for time-series outlier detection—demonstrating an ability to move ideas from papers into robust code. Known for improving environment fidelity and CPU compatibility in RL projects, he combines deep academic training with hands-on engineering that tightens the loop between research prototypes and production-ready pipelines.
9 years of coding experience
6 years of employment as a software developer
Bachelor's degree Computer Science, Bachelor's degree Computer Science at Wuhan UniversityDoctor of Philosophy - PhD Computer Science, Doctor of Philosophy - PhD Computer Science at Rice UniversityDoctor of Philosophy - PhD Student Computer Science, Doctor of Philosophy - PhD Student Computer Science at Texas A&M University
Github Skills (11)
machine-learning10
time-series10
python10
reinforcement-learning10
game-ai10
outlier-detection10
automl9
pytorch9
scikit8
scikit-learn8
git5
Programming languages (5)
Github contributions (5)
kwai/DouZero
Jun 2021 - Jun 2022
[ICML 2021] DouZero: Mastering DouDizhu with Self-Play Deep Reinforcement Learning | 斗地主AI
Role in this project:
ML Engineer
ML EngineerContributions:3 releases, 79 commits, 4 PRs in 1 year
Contributions summary:Daochen primarily contributed to the project by modifying the `env.py` file to include functionalities related to different objectives (wp, adp, and logadp) within the DouDizhu environment. They updated comments and added new features to the game environment, including reward calculations. Further, the user made updates to various setup files, including dependencies such as `rlcard`. The user also refined CPU support, enhancing the project's compatibility.
game-aiactor-criticreinforcement-learningdeep-reinforcement-learningmastering
datamllab/tods
Sep 2020 - Aug 2021
TODS: An Automated Time-series Outlier Detection System
Role in this project:ML Engineer
ML EngineerContributions:22 commits, 12 pushes, 1 comment in 11 months
Contributions summary:Daochen primarily contributed to the `BruteForceSearch` class, a core component for time-series outlier detection within the TODS system. Their work involved fixing bugs in the entry point and adding a brute-force search algorithm. They modified the pipeline to include different processing steps, feature analysis, and detection algorithms related to time-series analysis. They refined the example scripts and updated the pipeline structure within the AutoML system.
time-series-analysisoutlier-detectionanomaly-detectionmachine-learningoutlier
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.