AI Data Specialist (Confidential Global Project)
Executed high-precision multimodal labeling for text, image, and video data sets to train and evaluate Large Language Models on reasoning and safety. Routinely performed side-by-side (SxS) ranking and evaluation tasks to detect logical fallacies and "hallucinations" in AI-generated output. Used structured rubrics to validate factuality, compliance, and alignment with global standards. • Consistently achieved 100% compliance with meticulous data integrity standards • Applied RLHF and AI safety practices in daily annotation and evaluation • Leveraged expertise in bridging technical evaluations with user expectations • Ensured safe, accurate, and contextually sound model outcomes