AI Trainer & RLHF Specialist (Freelance)
Text
RLHFAs an AI Trainer and RLHF Specialist for Outlier AI and Handshake AI, I engaged in rating, ranking, and providing structured feedback on AI-generated responses to enhance model alignment and safety. I evaluated outputs from large language models (LLMs), covering areas such as creative content, coding, and reasoning, using annotation guidelines for high accuracy. This role leveraged my digital design and UI/UX expertise for domain-specific annotation and involved managing multiple annotation workflows in a remote setup. • Performed RLHF tasks for 6–12 months, providing preference rankings on AI outputs. • Identified errors in creative, design, coding, and reasoning outputs from LLMs for quality improvement. • Maintained accuracy by following diverse annotation guidelines across various tasks. • Reliably met project deadlines and quality thresholds in a freelance capacity.
Bounding Box