The company is looking for a freelance AI trainer to evaluate and benchmark advanced agentic audio models through role-play customer service scenarios. Candidates must have strong English verbal skills, basic programming literacy (JSON and logic), and high-quality audio equipment. The role involves assessing model performance based on conversational naturalness, task accuracy, and technical coherence.