AI Trainer & LLM Evaluator (Freelance)
As an AI Trainer & LLM Evaluator, I created and refined high-quality prompts and responses to train large language models. I evaluated and ranked AI-generated outputs using rubric-based criteria for accuracy, coherence, tone, and adherence to safety guidelines. I conducted fact-checking and research to ensure the factual reliability of training data and delivered actionable feedback to improve model performance. • Applied structured evaluation rubrics to maintain quality and consistency across tasks. • Identified model weaknesses, inconsistencies, and edge cases for ongoing improvement. • Provided detailed, project-aligned feedback to guide development cycles. • Supported training and evaluation across multiple domains in Brazilian Portuguese and English.