We are a series A start-up backed by Tier 1 investors including Foundry.vc and Next Frontier Capital.
If you have applied for an expert role within the past 12 months, you do not need to reapply.
About the role
We're seeking experienced radiologists to help evaluate the next generation of multimodal AI systems by assessing how they interpret medical images, radiology reports, and complex clinical instructions. This benchmark dataset project focuses on evaluating AI models for visual document understanding and instruction-following in the radiology domain. You'll develop challenging, clinically realistic evaluation tasks, create expert reference answers, and identify where today's AI systems succeed—and where they fall short.
What You'll Do
Design clinically realistic evaluation tasks using radiology images, reports, and supporting clinical documentation.
Create expert-level prompts that assess visual document understanding, diagnostic reasoning, and instruction-following across radiology workflows.
Evaluate AI-generated responses using structured rubrics for diagnostic accuracy, completeness, safety, and adherence to instructions.
Identify challenging edge cases, subtle imaging findings, and common model failure modes across multiple imaging modalities.
Provide detailed written feedback that helps improve model performance and benchmark quality.
Collaborate directly with research teams at a leading AI lab to advance multimodal healthcare AI.
Participate in onboarding, calibration sessions, and ongoing quality reviews.
Qualifications
MD, DO, or internationally equivalent medical degree.
Completion of an accredited Diagnostic Radiology residency (ACGME, AOA, or recognized international equivalent).
Current board certification in Diagnostic Radiology (ABR, FRCPC, FRANZCR, or recognized national equivalent).
Current, active, unrestricted medical license with no suspension, revocation, or ongoing disciplinary action.
At least 3 years of independent clinical practice following residency (training years excluded), currently practicing or having practiced within the past two years.
Experience practicing in an accredited or government-licensed healthcare setting, including hospitals, academic medical centers, outpatient imaging centers, or private practice.
Nice to have
Currently providing at least 10 hours of direct clinical interpretation per week.
Fellowship training or significant clinical expertise in a radiology subspecialty, including:
Neuroradiology
Musculoskeletal Radiology
Body Imaging
Breast Imaging
Cardiothoracic Radiology
Pediatric Radiology
Interventional Radiology
Emergency Radiology
Experience interpreting multimodal imaging studies, including CT, MRI, ultrasound, mammography, nuclear medicine, and plain radiography.
Experience reviewing AI-assisted imaging tools or participating in imaging AI research is a plus.
Passion for advancing trustworthy AI in healthcare.
Compensation
This is a contract, project-based position. The hourly rate will depend on experience and project parameters. The rate will be disclosed prior to starting the project.
IMPORTANT - How to Apply
We can only accept and review and applications via our direct official recruiting portal: https://mpathic2.bamboohr.com/careers, due to the abundance of 3rd party apps and automated agents. Only applications submitted directly by the applicant through our portal will be considered.
All Job Ads are subject to GrabJobs’s Terms of Service. We allow users to flag postings that may be in violation of those terms. Job Ads may also be flagged by GrabJobs moderation team. However, no moderation system is perfect, and flagging a posting does not ensure that it will be removed.
Be the first to receive the latest Others Full-Time Jobs in the US.
Setup your job alert:
By activating job alerts, I agree to GrabJobs Terms & Privacy Policy. I can unsubscribe to job alerts anytime.
Skip
GrabJobs is the no1 job portal in the US, connecting you to thousands of jobs fast!
Find the best jobs in the US, apply in 1 click and get a job today!