Cosmology AI Benchmark Designer

Mercor; Greater London; Remote; GBP 41,000 - 83,000; Full time

Job summary

Mercor is seeking a Computational Astrophysics & Cosmology Expert to design graduate-level problems that test AI systems using real scientific software. You will create challenging computational tasks, run simulations, interpret results, and refine problems across AI models.

Requirements include MS/PhD in a relevant STEM field, hands-on proficiency with libraries such as Astropy, strong Python skills, and ability to work in Linux/remote compute sandboxes for 15–20 hours per week.

Qualifications

• Graduate-level training in a STEM field (MS/PhD preferred).
• Proven hands-on use of scientific software libraries (e.g., Astropy) in research or publications.
• Strong Python skills for writing problem setups, oracle functions, and validators.

Responsibilities

• Design graduate-level computational problems based on real scientific workflows.
• Test problems against state-of-the-art AI models and refine until target difficulty is met.
• Interpret results, plan experiments, and ensure reproducibility in sandboxed environments.

Skills

Python

Astropy

Cosmology

Research experience

Education

MS/PhD in relevant STEM field

Tools

Linux/Unix

Job description

Mercor is seeking a Computational Astrophysics & Cosmology Expert to design graduate-level problems that test AI systems using real scientific software. You will create challenging computational tasks, run simulations, interpret results, and refine problems across AI models.

Requirements include MS/PhD in a relevant STEM field, hands-on proficiency with libraries such as Astropy, strong Python skills, and ability to work in Linux/remote compute sandboxes for 15–20 hours per week.