AI Trainer & Data Labeler (Freelance) – OpenTrain.ai / Self-driven projects
As an AI Trainer and Data Labeler, actively participated in RLHF, Response Evaluation, Safety Alignment, and Red Teaming for large language models. Utilized legal expertise to annotate, audit, and review legal and compliance-related texts including contracts, policies, and cryptocurrency documentation. Used Label Studio and Scale AI to structure high-quality, instruction-following datasets for LLM training. • Focused on legal compliance assessment and risk labeling. • Contributed to safety, harmlessness, and helpfulness reviews for conversational AI. • Evaluated, classified, and annotated complex legal and blockchain texts. • Ensured precise multi-turn dialogue and edge case identification.