Physics LLM Evaluation Expert

You will help evaluate and improve large language models by creating challenging physics and STEM problems. Your work will test model reasoning, symbolic manipulation, abstraction, and ability to explain complex ideas clearly.

• Design and solve difficult STEM and physics problems that reveal model limitations.
• Write accurate, step-by-step solutions with clear reasoning.
• Work with LLM researchers to align problems with evaluation goals.
• Help create evaluation benchmarks based on physics curricula from early undergraduate through PhD level.
• Break down complex physics concepts into simple, precise explanations.
• Provide detailed annotations and constructive feedback.

What It Pays And Takes

This is an expert-level, part-time contractor role for someone who can work independently in a remote setting. The listing does not provide a payment rate.

• Hours: 20 or more hours per week.
• Location: Worldwide and remote.
• Language: Excellent English comprehension and structured communication are required.
• Experience: Graduate, PhD, or postdoctoral-level STEM study or experience is expected.
• Background: Strong physics, mathematics, and related STEM knowledge.
• Helpful experience: Applied physics or a related STEM specialization.
• Helpful ability: Explain advanced physics concepts using simple language, visuals, and clear reasoning.
• Work style: Strong analytical and research skills, detailed annotation ability, and independent working habits.

How It Works

Apply on OpenTrain with your resume, then complete the application on the hiring site.

About AI Training Work

AI training is the human work behind systems that generate and understand text, images, code, and other content. Contributors create examples, review model responses, and test how well models reason, which is why deep subject knowledge matters for expert evaluation work.