Data Labeling Pipeline overview

Manages the end-to-end data labeling workflow — task design, labeler onboarding, quality assurance sampling, inter-annotator agreement measurement, active-learning-driven sample selection, and labeled dataset versioning. Excludes model training.