Ailing Zeng is a researcher with eight years of experience building multimodal, human-like intelligent agents on scalable big data, currently leading multimodal video generation and understanding at Anuttacon. She has steered research groups at Tencent and the International Digital Economy Academy, focusing on controllable video generation, novel architectures, and multimodal interactions and evaluation. Her academic foundation includes a PhD in Computer Science from The Chinese University of Hong Kong and multiple engineering and mathematics degrees that underpin a strong cross-disciplinary approach. She contributes to applied open-source work—having adjusted and tuned models and training pipelines for the NeurIPS 2022 SCINet time-series repo—showing hands-on expertise from research design to code-level experimentation. Colleagues describe her as a tech lead who blends deep academic rigor with practical system-building and hyperparameter-driven experimentation. She often explores edge cases in human-centric perception, pushing beyond typical benchmarks to improve real-world multimodal fidelity.
8 years of coding experience
3 years of employment as a software developer
Bachelor of Engineering - BE, Computer Systems Networking and Telecommunications, Bachelor of Engineering - BE, Computer Systems Networking and Telecommunications at 中山大学
The GitHub repository for the paper: “Time Series is a Special Sequence: Forecasting with Sample Convolution and Interaction“. (NeurIPS 2022)
Role in this project:
Back-end Developer
Contributions:92 commits, 45 pushes, 9 comments in 1 year
Contributions summary:Ailing primarily updated Python scripts (`.py` files), specifically focusing on modifying argument parsing within the `run_ETTh.py` and `run_financial.py` scripts and adjusting parameters for models used within the `cure-lab/scinet` repository. These changes involve adjusting settings related to dataset configuration, device utilization (GPU), input/output sequence lengths, and training parameters. This indicates the user was involved in experimenting with different configurations and datasets related to time series forecasting.
Code for "SRNet: Improving Generalization in 3D Human Pose Estimation with a Split-and-Recombine Approach" ECCV'20
Contributions:21 commits, 8 pushes, 1 branch in 3 months
pytorchhuman-posesplitgeneralizationapproach
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.