Zhi Li
Machine Learning Engineer at Netflix
Los Gatos, California, United States
Summary
🤩
RockstarZhi Li is a Machine Learning Engineer based in Los Gatos, California with 10 years of experience applying signal processing and ML to multimedia and networking problems. He has hands-on expertise in perceptual video quality assessment, contributing backend and data science improvements to Netflix's widely used VMAF project, including dataset readers, noisy-image handling, and advanced training/metrics pipelines. Zhi bridges research-grade algorithms and production systems, focusing on compression-aware video analytics and robust feature engineering. He brings a practical, metrics-driven approach to model evaluation and integration, often augmenting core frameworks rather than rebuilding them. Colleagues rely on him to translate complex statistical methods into scalable tooling that improves real-world video quality assessment.
10 years of coding experience
Github Skills (18)
data-analysis10
python10
data-science10
video-processing10
machine-learning9
performance-tuning8
apidoc8
api8
performance-analysis8
performance-measurement8
performance-analytics8
performance-monitor8
performance-monitoring8
image-processing7
architectures6
Programming languages (3)
Github contributions (5)
Netflix/vmaf
Jan 2016 - Jan 2023
Perceptual video quality assessment based on multi-method fusion.
Role in this project:
Back-end Developer & Data Scientist
Back-end Developer & Data ScientistContributions:16 releases, 54 reviews, 1328 commits in 7 years
Contributions summary:Zhi's contributions primarily focused on enhancing the perceptual video quality assessment framework (VMAF). They implemented new dataset readers for both standard and noisy image datasets, including the addition of various features to the training process, such as, the enabling of a scoring process for different model types and the addition of methods for measuring statistical metrics to analyse the performance and accuracy of the framework. They also integrated with and expanded the capabilities of the existing core functionalities.
pytorchvideo-quality-assessmentfeature-detectionperceptualmethod
Netflix/vmaf_resource
Apr 2020 - Sep 2022
Contributions:9 commits, 7 pushes in 2 years 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.