AI Data Trainer & Prompt Engineer, Outlier
As an AI Data Trainer & Prompt Engineer at Outlier, I specialized in training large language models using reinforcement learning from human feedback. My core responsibilities involved authoring, refining, and rating complex prompts across domains such as coding, logic, and creative writing, as well as performing data annotation and dataset evaluation to enhance model reasoning and performance. I conducted rigorous quality assurance and fact-checking on AI-generated content to ensure factual accuracy, safety, and adherence to guidelines. • Developed and reviewed prompt-engineering tasks for LLMs. • Annotated and evaluated datasets for continual improvement. • Performed QA and fact-checking of model outputs for policy compliance. • Collaborated remotely with a continuous workflow over two years.