AI Training & Quality Assurance Specialist
As an AI Training & Quality Assurance Specialist at Appen/CrowdGen and Outlier, I improved AI models using Reinforcement Learning from Human Feedback (RLHF) and data annotation. My responsibilities included verifying peer-submitted data for accuracy, auditing outputs against detailed guidelines, and evaluating multi-turn conversations. I leveraged prompt engineering and advanced instruction mastering to ensure outputs met high standards of safety, alignment, and professionalism. • Promoted to Lead Auditor to ensure 95%+ adherence to extensive style guides. • Evaluated and ranked AI-generated conversations across technical, legal, and creative domains. • Designed and stress-tested complex prompt structures for logic, empathy, and accuracy. • Conducted human tone evaluations to eliminate non-professional AI responses.