Dermatologist to Enhance and Evaluate LLM and AI

• A minimum of 5 years of post-residency experience practicing as a dermatologist
• he work involves assessing image quality and diagnostic usability
• Review AI-generated medical text for accuracy, harmfulness, and completeness, with focus on dermatology.

Role Overview:

We are seeking an experienced Dermatologist to help evaluate and enhance large language model (LLM) outputs in the medical domain. You’ll assess AI-generated clinical content for dermatological accuracy, safety, and relevance — contributing to the development of responsible AI in healthcare.

We are seeking US-based dermatologists to review dermatology images using a structured online evaluation tool. The work involves assessing image quality and diagnostic usability; identifying concerns such as blur, lighting, exposure, colour, and framing; and, where applicable, identifying likely skin conditions, assigning confidence ratings, and documenting rationale.
 

What You'll Do Day-to-Day:

• Review AI-generated medical text for accuracy, harmfulness, and completeness, with focus on dermatology.

• Evaluate diagnoses, treatment plans, and summaries for dermatologic cases.

• Provide expert feedback to improve medical integrity and safety of LLM outputs.

• Collaborate with AI researchers and clinicians to align model performance with evidence-based dermatology standards.

Requirements:

• Based in the US (mandatory).

• Licensed Medical Doctor (MD or MBBS) with specialization in Dermatology or equivalent qualification.

• Must be fully licensed and practicing medicine within the USA.

• Minimum 5 years post-residency experience practicing as dermatologist.

• Strong visual attention to detail

• Comfortable reviewing sensitive or graphic medical images

• Technically savvy—able to navigate web-based annotation platforms, use image zoom/tools, follow detailed guidelines, and accurately complete structured questionnaires.

• Experience in medical content review, clinical research, annotation or AI/ML evaluation preferred.