Quantum Computing Scientist AI Benchmarking & Prompt Design

Obsidian; Greater London; Remote; GBP 7,200 - 11,000; Part time

Job summary

Mercor in the United Kingdom is seeking PhD and Master's physicists to author AI evaluation tasks for a new frontier benchmark. You will create original, executable problems that current models struggle to solve, focusing on at least two physics subdomains and leveraging Python or R for computation.

This six-week, part-time engagement (20+ hours/week) starts immediately, with opportunities to publish and contribute to a growing AI benchmarking effort.

Qualifications

• PhD in physics or related field is required.
• Depth in at least two subdomains (e.g., condensed matter, optics, quantum information).
• Proficient in Python, R, or other scientific computing languages; familiar with Git workflow and Docker.

Responsibilities

• Source your own material (papers, datasets, open-source repos) or design scenarios.
• Write scientific prompts based on inputs.
• Build grading criteria to define correct answers.
• Calibrate against frontier models; a task ships when models fail it more often than succeed.

Skills

Python

R

Scientific computing

Git/GitHub

Docker

Education

PhD in physics

Master's in physics

Job description

Mercor in the United Kingdom is seeking PhD and Master's physicists to author AI evaluation tasks for a new frontier benchmark. You will create original, executable problems that current models struggle to solve, focusing on at least two physics subdomains and leveraging Python or R for computation.

This six-week, part-time engagement (20+ hours/week) starts immediately, with opportunities to publish and contribute to a growing AI benchmarking effort.