Legal AI Trainer
As a Legal AI Trainer, I was responsible for evaluating and reviewing AI-generated legal document outputs for quality and accuracy. My work included assessing model responses for legal soundness, reasoning, and logical consistency across multiple domains. I followed detailed project guidelines and provided structured feedback to improve AI model performance. • Conducted evaluations for outputs in real estate, contract, criminal, and regulatory law. • Identified hallucinations, unsupported legal conclusions, and reasoning errors in AI responses. • Applied structured annotation and consistent evaluation standards across assignments. • Provided written justifications for decision-making in alignment with project requirements.