Tao Liu
Summary
🤩
RockstarTao Liu is an AI research engineer with a PhD in Computer Vision from BUPT and four years of industry experience building production-ready multimodal and vision systems. Currently at Centific, he has driven automated multi-agent evaluation to improve annotation quality, cut review effort, and used RL fine-tuning to replace large models with optimized small ones for lower serving cost. His prior work includes high-fidelity 3D human face reconstruction, text-to-motion synthesis (ECCV 2024), and deploying super-resolution models to Snapdragon NPUs for Android. He blends deep research experience—from a visiting scholar stint at UW on object detection and tracking—with pragmatic engineering that ships benchmarks and mobile deployments. Also active in audio-visual research (speaker diarization and talking-face generation), he combines end-to-end ML training with systems know-how to move models from lab to device. Notably, he requires no immigration sponsorship and is based in Kent, Washington.
4 years of coding experience
Github Skills (94)
text-to-speech10
natural-language-generation10
chatbot10
chatgpt9
video-understanding9
fine-tuning9
high-resolution9
python9
nlp8
ava8
animator8
compression8
gmm8
action-recognition8
mixture-of-experts8
Programming languages (3)
Github contributions (5)
X-LANCE/MSDWILD
Mar 2022 - Nov 2022
Contributions:17 commits, 30 pushes, 3 branches in 8 months
liutaocode/AwesomeDiarizationDataset
Nov 2022 - Dec 2022
Contributions:18 commits, 16 pushes, 1 branch in 29 days
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.