Radiologist for AI Model Evaluation

Help establish the clinical reference standards used to evaluate advanced AI systems on real radiology studies. This role draws on the diagnostic reasoning that matters most in practice, including correct measurement conventions, comparison with prior imaging, and structured classifications in challenging cases. It is not image labeling.

• Review de-identified CT, MR, radiography, ultrasound, PET/CT, mammography, and other imaging studies, then produce a reference report or targeted finding.
• Apply practice-based structured systems and measurement conventions, including RECIST, Lung-RADS, BI-RADS, CAD-RADS, ASPECTS, and others.
• Assess AI-drafted interpretations and identify findings that are inaccurate, unsupported, or unsafe.
• Contribute challenging cases that test the limits of strong AI models.

• Board-certified or board-eligible radiologist, with ABR, FRCR, EDiR, or an international equivalent.
• Later-stage residents and fellows may be considered.
• Diagnostic radiology experience in any subspecialty, including neuro, chest, abdominal, MSK, breast, cardiac, nuclear medicine, paediatric, or others.
• Currently interpreting studies or recently active in clinical practice.

• Remote, hourly engagement.

• $300 to $400 per hour.