Senior AI Trainer & RLHF Specialist
As a Senior AI Trainer and RLHF Specialist at Outlier AI (Scale AI Partner), I evaluated and ranked LLM model outputs for usefulness, accuracy, and safety. I created high-quality preference rankings and conducted red-teaming to uncover model vulnerabilities. My labeling work supported reward model pipelines and contributed to measurable model improvements. • Ranked LLM responses using RLHF for finance, business, and technology prompts. • Delivered consistent, high-quality preference annotation across 2,000+ tasks. • Performed red-teaming to surface hallucinations and policy violations. • Maintained above 95% quality score across all assignments.