This remote, entry-level contract position involves evaluating AI-generated responses to social scenarios and providing written explanations for judgments to improve model behavior. The project runs for 2 to 3 weeks, requires at least 20 hours per week, and is open to applicants worldwide. Candidates need strong written English skills, ideally with a background in business analysis, writing, journalism, literature, or content evaluation.