Materials Scientist or Engineer for AI Model Evaluation

Design rigorous, real-world materials science and engineering tasks that evaluate an AI model’s expert reasoning. You will create prompts, supporting data rooms, and grading criteria, then test and refine each task until it clearly distinguishes strong from weak performance.

• Create domain-authentic tasks based on your own professional experience.
• Develop the data room and objective scoring criteria for each task.
• Run tasks against the model, analyze successes and failures, and refine tasks until they meaningfully discriminate performance.
• Partner with the project lead and fellow subject-matter experts to define high-quality work in your specialty.

• Senior-level expertise in materials science, including metals, ceramics, composites, characterization, or additive manufacturing, or in electrical engineering, including analog, RF, or power, or mechanical engineering, including design, CFD and fluids, heat transfer, pressure vessels, NDE, or CAE.
• Comfort working through ambiguity and helping shape an evolving body of work.
• Ability to explain both why an answer is correct and why an incorrect answer fails.

• Remote, hourly engagement.

• $60 to $90 per hour.

Daily onboarding sessions and multiple office hours are available each day to support your start and ongoing work.