AI Evaluator / RLHF Rater (Freelance)
As an AI Evaluator and RLHF Rater at Outlier AI, I assessed, edited, and rated responses generated by large language models in Brazilian Portuguese. My role focused on verifying factual accuracy, cultural suitability, and ensuring the safety of AI-generated content. I refined texts to guarantee impeccable grammar, logical coherence, and a natural, humanized tone.• Conducted thorough evaluations of LLM outputs for factual consistency and appropriateness • Edited and corrected grammar, style, and coherence in Portuguese texts • Applied RLHF methodologies for AI refinement and improvement • Ensured content quality and safety for end users