• Design and role-play chat experiences with AI agents across diverse clinical and emotional scenarios • Perform or simulate characters across ages, backgrounds, severity levels, and emotional states (spoken or written) • Write dialogue scripts, scenario outlines, or narrative prompts for testing edge cases in mental-health contexts • Conduct qualitative analyses of conversations to derive taxonomies, personas, and behavioral patterns • Translate clinical expertise into structured prompt patterns and evaluation rubrics • Provide expert clinical feedback on internal mental-health policy development • Collaborate with engineering and research teams to define evaluation metrics for tone, psychological validity, and appropriateness • Identify and document failure cases, risk signals, and edge behaviors • Contribute to scenario modeling, red teaming, and rapid experimentation cycles • Ensure all work adheres to strict confidentiality agreements and NDAs • Implement quality-assurance protocols for conversation and behavioral analysis • Participate in review sessions with engineers, researchers, and clinical consultants