Particle Physics Problem Designer for AI Benchmarks

Obsidian; New York (NY); On-site; USD 83,000 - 138,000; Part time

Job summary

Obsidian is seeking a Particle & Nuclear Physics Expert to design graduate-level problems that require real scientific software, run simulations, interpret results, and design experiments. You will work with scikit-hep and related HEP Python tools to analyze data, perform cross-section calculations, and explore perturbative QCD concepts.

The role emphasizes puzzle-driven problem design over mere computation, with hands-on coding to solve authentic research problems.

Qualifications

• Graduate-level training in a relevant STEM field (MS, PhD, or equivalent research experience)
• Proven proficiency with at least one scientific software library (e.g., scikit-hep) and related tools
• Strong Python skills — you'll be writing problem setups, oracle functions, and solution validators
• Ability to work independently and refine problem designs based on feedback
• Comfortable working in a Linux/terminal environment with remote compute sandboxes
• Available for at least 15–20 hours per week

Responsibilities

• Design original graduate-level problems based on real scientific workflows
• Run simulations and interpret results using specialized scientific software
• Test problems against AI models and refine until target difficulty is reached
• Collaborate with the project team to tune difficulty and ensure reproducibility

Skills

Python programming

Linux/Unix

Puzzle-driven design

Education

MS/PhD in Physics or related STEM field

Tools

scikit-hep

Monte Carlo tooling

Job description

Obsidian is seeking a Particle & Nuclear Physics Expert to design graduate-level problems that require real scientific software, run simulations, interpret results, and design experiments. You will work with scikit-hep and related HEP Python tools to analyze data, perform cross-section calculations, and explore perturbative QCD concepts.

The role emphasizes puzzle-driven problem design over mere computation, with hands-on coding to solve authentic research problems.