AI Advanced Data Trainer | Invisible Technologies
Text
RLHFOptimized machine learning models using Reinforcement Learning with Human Feedback (RLHF) and Supervised Fine-Tuning (SFT) workflows for NLP. Annotated, labeled, and audited complex NLP datasets to improve conversational accuracy and safety constraints. Validated data accuracy and evaluated model outputs with research verification and citations for responses. • Created and refined training data for STEM-based NLP model behavior • Performed dataset labeling and quality auditing to ensure technical accuracy • Conducted output evaluation and verification with citation support • Calibrated training operations to reinforce safety constraints
2024 - Present