Description

Job DescriptionExpert Contributor - Radiologist

Task-Based Independent Contractor Opportunity

Overview

We're looking for Radiology experts who can design and solve challenging, real-world reasoning tasks that test the limits of advanced AI systems. You'll create high-quality radiology problems, develop rigorous solutions, and help identify exactly where models fall short in clinical and diagnostic reasoning. The focus is on graduate-level difficulty, diagnostic precision, and reproducible evaluation. If you thrive on building nuanced diagnostic scenarios and evaluating complex imaging interpretations, this opportunity is calling your name.

Key Responsibilities

• Design complex, multi-step diagnostic reasoning problems with precise, fully worked interpretations and clinical decision pathways.
• Evaluate AI-generated diagnostic reasoning and identify where models produce incorrect or incomplete analyses.
• Create problems spanning CT, MRI, X-ray, ultrasound, and nuclear medicine imaging modalities.
• Develop reproducible evaluation criteria and grading rubrics for each diagnostic scenario.
• Review and critique competing diagnostic conclusions across multi-turn evaluation sessions.
• Collaborate asynchronously with the project team on clinical accuracy and difficulty calibration.

Required Qualifications

• Deep fluency in Radiology, with the ability to create complex, multi-step diagnostic reasoning problems and produce precise, fully worked interpretations.
• Proficiency with imaging modalities including CT, MRI, X-ray, ultrasound, and nuclear medicine, as well as PACS and relevant diagnostic software.
• Strong review instincts with the ability to identify subtle diagnostic errors and evaluate competing conclusions.
• MD with Radiology residency completion, fellowship, or equivalent; board-certified or board-eligible candidates strongly preferred.

Preferred Qualifications

• Background in academic radiology, residency/fellowship instruction, case-based curriculum design, or clinical research.
• Familiarity with board exam-style question design (ABR, FRCR, or equivalent).
• Experience evaluating AI-generated diagnostic reasoning or working with AI-assisted imaging tools.
• Published research in a peer-reviewed radiology or medical imaging journal.