AI Data Annotator/Trainer
As an AI Data Annotator/Trainer, I was responsible for labeling and curating high-quality datasets across computer vision, NLP, multimodal, and LLM fine-tuning projects. I consistently achieved high annotation accuracy and strong inter-annotator agreement scores. My work included following complex guidelines and identifying edge cases to ensure data quality. • Labeled 5,000+ data points including images, text, and multimodal inputs. • Performed RLHF-style preference ranking, safety evaluations, and feedback on LLM outputs. • Maintained daily throughput of up to 3,000 annotations/day with high-quality standards. • Utilized tools such as Labelbox, CVAT, SuperAnnotate, Scale AI, Label Studio, Prodigy, Amazon SageMaker Ground Truth, VGG Image Annotator, and Snorkel.