Computational Chemist: AI Benchmark Developer (6-Week)

Mercor; Greater London; Remote; GBP 6,612,000 - 9,919,000; Part time

Job summary

Mercor is seeking PhD and MSc scientists to author AI evaluation tasks (Sci Code) for a new benchmarking effort in scientific computing. You will design original, executable research problems in quantum and computational chemistry, source materials, and craft prompts with robust grading criteria.

The role emphasizes testing frontier models’ capabilities and improving evaluation pipelines. Engagement runs for 6 weeks, part-time (20+ hours/week), with immediate start.

Qualifications

• PhD in chemistry or closely related field required.
• Demonstrated depth in both quantum chemistry and computational chemistry.

Responsibilities

• Author AI evaluation tasks and scientific prompts.
• Develop grading criteria for answers.
• Calibrate prompts against frontier models to identify challenging cases.
• Source and curate material (papers, datasets, repositories) for tasks.

Skills

Quantum chemistry

Computational chemistry

Python

R

Scientific programming

Education

PhD in chemistry

Master's in chemistry

Tools

Git/GitHub

Docker

Job description

Mercor is seeking PhD and MSc scientists to author AI evaluation tasks (Sci Code) for a new benchmarking effort in scientific computing. You will design original, executable research problems in quantum and computational chemistry, source materials, and craft prompts with robust grading criteria.

The role emphasizes testing frontier models’ capabilities and improving evaluation pipelines. Engagement runs for 6 weeks, part-time (20+ hours/week), with immediate start.