Materials Science PhD — AI Benchmark Architect

Mercor; New York (NY); On-site; USD 83,000 - 124,000; Part time

Job summary

Mercor is hiring PhD and Master's scientists to author AI evaluation tasks for Sci Code. You will craft original, executable research problems that frontier models cannot solve.

Domains include materials science with a focus on semiconductor materials and molecular modeling. You will source material, write prompts, define grading criteria, and calibrate against models, ensuring tasks challenge state-of-the-art systems.

Qualifications

• PhD in materials science, materials engineering, applied physics, chemistry, or related field.
• Depth in semiconductor materials and molecular modeling.
• Proficiency in Python, R, or similar scientific computing languages.
• Experience with Git/GitHub and Docker in PR-driven workflows.

Responsibilities

• Source your own material from papers, datasets, or open repositories.
• Write scientific prompts based on inputs.
• Develop grading criteria for correct answers.
• Calibrate prompts against frontier models to fail cases.

Skills

Python

R

Git/GitHub

Docker

Education

PhD in Materials Science

Master's degree in related field

Tools

Docker

GitHub

Job description

Mercor is hiring PhD and Master's scientists to author AI evaluation tasks for Sci Code. You will craft original, executable research problems that frontier models cannot solve.

Domains include materials science with a focus on semiconductor materials and molecular modeling. You will source material, write prompts, define grading criteria, and calibrate against models, ensuring tasks challenge state-of-the-art systems.