This remote contractor role involves reviewing and evaluating AI text responses in human-AI social scenarios to improve model behavior. Candidates need intermediate-level experience in data annotation or analytical content evaluation, strong written English proficiency, and objective judgment in nuanced interpersonal situations. The project runs for 2 to 3 weeks and requires a commitment of at least 20 hours per week.