Data Annotation (Handshake AI)
Performed RLHF (Reinforcement Learning from Human Feedback) data annotation tasks on the Handshake AI platform, evaluating and ranking LLM-generated conversational responses for quality, helpfulness, accuracy, and safety. Reviewed chatbot outputs across diverse general knowledge prompts, providing structured preference rankings and written rationales to support reward model training. Adhered strictly to project guidelines and quality benchmarks throughout the engagement.