Remote Diagnostic Radiologist AI Model Evaluator Part-Time

Mercor; New York (NY); Remote; USD 165,000 - 331,000; Part time

Job summary

Mercor is hiring board-certified radiologists to evaluate health AI models in partnership with a leading AI lab. You will read de-identified patient charts and judge AI answers to clinical questions in radiology.

Remote, hourly, with flexible part-time hours; start date immediate. Required: MD/DO, ABR Diagnostic Radiology certification, active US license, and 1+ year of independent practice. 3+ years post-training preferred.

Qualifications

• MD or DO from an accredited medical school.
• Board certified in ABR Diagnostic Radiology through an ABMS member board or the AOA.
• Active, unrestricted US medical license.
• Residency complete, plus fellowship where your specialty requires one.
• Currently practicing in the US with at least 1 year of independent practice.
• 3+ years of post-training experience strongly preferred.
• Clear written communication and close attention to detail.

Responsibilities

• Evaluate AI answers to radiology questions for clinical accuracy, safety and alignment with current standards of care
• Check each answer against the patient chart and decide which of two responses is stronger
• Write structured rationales on clinical rigor, completeness and clarity
• Flag errors that could compromise patient safety

Skills

MD/DO

ABR Radiology board

US medical license

Attention to detail

Education

MD or DO from an accredited medical school

Board certified in ABR Diagnostic Radiology

Active, unrestricted US medical license

Job description

Mercor is hiring board-certified radiologists to evaluate health AI models in partnership with a leading AI lab. You will read de-identified patient charts and judge AI answers to clinical questions in radiology.

Remote, hourly, with flexible part-time hours; start date immediate. Required: MD/DO, ABR Diagnostic Radiology certification, active US license, and 1+ year of independent practice. 3+ years post-training preferred.