Chatbot Training & Evaluation Project
I trained and evaluated conversational AI models by ranking chatbot responses and providing structured reinforcement learning from human feedback (RLHF). My work included identifying intents, classifying conversations, and improving chatbot safety and relevance. This role required prompt evaluation and detailed tagging to optimize model performance. • Ranked and rated chatbot responses for conversational AI. • Applied RLHF methods for model improvement. • Identified and flagged harmful outputs in dialog data. • Collaborated to refine annotation guidelines and quality.