Key Responsibilities
• Create and refine high-quality training datasets for NLP and computer vision models. • Apply annotation guidelines compliance across tasks: NER, sentiment, intent, classification, bounding boxes, polygons, segmentation, OCR, audio transcription. • Conduct QA evaluation, spot checks, inter-annotator agreement (IAA), error taxonomy, and remediation. • Perform RLHF workflows: rank responses, judge instruction following, rate safety/harms, and conduct large language model evaluation. • Execute prompt evaluation and adversarial testing for LLM safety and robustness. • Document edge cases, update labeling SOPs, and provide feedback to improve model performance. • Collaborate with project managers, ML engineers, and QA leads via Rex.zone tooling.



