At Outlier AI, I specialized in image and video annotation for computer vision model training, delivering guideline-compliant labels across diverse datasets.
I performed 2D/3D bounding boxes, semantic segmentation, polygon tracing, and frame-level video labeling.
I reviewed and corrected annotations before submission to maintain quality standards, and adapted to evolving project guidelines and annotation tools.
I worked independently to meet throughput and accuracy targets, and collaborated with project leads to resolve ambiguous cases and improve labeling guidelines.
