AI Trainer
Scope. Supported computer vision model training for object detection and segmentation across diverse datasets, including urban scenes, and everyday objects. Specific Data Labeling Tasks. Drew bounding boxes around objects (e.g., vehicles, people). Created polygon annotations for precise object segmentation. Applied class labels and attributes based on predefined taxonomies. Reviewed and corrected edge cases such as occlusions and overlapping objects. Project Size Annotated 5000+ images across multiple datasets. Worked in batch assignments with daily targets and weekly deliverables. Quality Measures Maintained annotation QA accuracy of not less than 95%. Followed strict annotation guidelines and validation rules. Conducted peer reviews and self-QA checks before submission. Incorporated feedback loops, reducing labeling errors by 30%.