Anh Duy Nguyễn Huỳnh

I'm a freelance video and multimodal data annotator working on POV and robotics footage for computer vision training.

I identify precise pick-and-place action boundaries, perform multi-class object labeling, track spatial orientation, and categorize piles with over 98% consistency.

Using CVAT, I create 2D bounding-box tracks and keyframe interpolation while maintaining Track ID continuity through occlusions and dynamic motion. I also evaluate audio and video datasets against strict scoring rubrics and document edge cases for QA review.

Previously, at Bonjour Resto’, I reviewed Vietnamese promotional content and scripts, managed daily communications, and updated customer feedback reports. I bring disciplined guideline adherence, strong visual discrimination, and native Vietnamese with professional English proficiency.