Specializing in Large Language Models, Computer Vision, and Deep Learning Optimization with 10+ years of experience.
Currently at TikTok Trust & Safety | Previously at Meta AI Research (FAIR) | CVPR Spotlight Author
I hold a Master's degree in Control Engineering from Tsinghua University, where I focused on deep learning optimization, human pose estimation, and face recognition. My research has been published at top venues including CVPR (Spotlight), IEEE TNNLS, and Pattern Recognition, and I hold a U.S. patent for automated checkout tracking systems.
With experience at TikTok, Meta FAIR, and AiFi, I've built production ML systems ranging from multimodal content moderation to RGB-only autonomous checkout solutions. I'm a peer reviewer for CVPR, ECCV, ICCV, and IEEE TPAMI. My expertise spans the full ML stack: from algorithm research to deployment with Docker, Kubernetes, and edge devices.
CVPR 2018 Spotlight paper: A novel approach applying PID control theory to stochastic optimization of deep networks. Extended to IEEE TNNLS with impact factor 11.368. This work bridges control engineering and deep learning optimization.
At TikTok: Building Large Language Model based ML/CV/Speech multimodal systems to identify policy-violating content at scale. Combining vision, language, and audio understanding for comprehensive content moderation.
At AiFi: Led computer vision detection algorithms for RGB-only auto-checkout solution (similar to Amazon Go). Developed customer detection + tracking and product detection + tracking systems. Resulted in U.S. Patent US11393213B2.
At HK PolyU: Developed end-to-end facial analysis system including face detection, alignment, and recognition. Research contributed to advances in human pose estimation and multi-person 2D pose estimation (ICME 2018).
San Jose, California, USA
Trust & Safety team: Prototyping Large Language Model based ML/CV/Speech multimodal systems to identify policy-violating content on the platform at scale.
September 2021 - PresentMenlo Park, California, USA
Facebook AI Research: Worked on Explainable AI to open the 'black box' of computer vision models, and Responsible AI for inclusive & fairness in Computer Vision systems.
December 2020 - August 2021Santa Clara, California, USA
Led computer vision detection algorithms for RGB-only autonomous checkout solution (similar to Amazon Go). Developed customer detection + tracking and product detection + tracking systems. U.S. Patent holder.
June 2018 - November 2020Hong Kong
Developed PID optimizer to accelerate CNN training (CVPR 2018 Spotlight). Built the PolyU-Face system for face detection, alignment, and recognition.
October 2016 - April 2018Beijing, China
Control Engineering with research focus on deep learning optimization, human pose estimation, object detection, and face recognition. Student instructor for Big Data class.
August 2015 - July 2018Taiyuan, China
Focused on tuning PID controller parameters in steel milling machines. Applied control theory to industrial automation.
July 2012 - September 2014Kunming, China
Instrument and Measurement Science with concentration in control theory. Foundation in PID control systems and signal processing.
September 2008 - July 2012