Medical Data Trainer
TextRLHF
• Evaluate AI-generated medical responses for clinical accuracy, safety, guideline adherence, and reasoning quality
• Identify hallucinations, unsafe recommendations, missed red flags, and flawed differential diagnoses
• Rank and benchmark model outputs across multiple clinical and biological task domains
• Provide structured feedback supporting model calibration and quality improvement
• Apply physician-level judgment to assess real-world clinical applicability and risk
• Maintain high consistency and quality standards in high-volume evaluation workflows
• Evaluate AI-generated medical responses for clinical accuracy, safety, guideline adherence, and reasoning quality • Identify hallucinations, unsafe recommendations, missed red flags, and flawed differential diagnoses • Rank and benchmark model outputs across multiple clinical and biological task domains • Provide structured feedback supporting model calibration and quality improvement • Apply physician-level judgment to assess real-world clinical applicability and risk • Maintain high consistency and quality standards in high-volume evaluation workflows
2026 - Present