Computational Chemist for AI Benchmark Design

Role Overview

Design graduate-level computational chemistry challenges that evaluate whether advanced AI systems can use scientific software for research-oriented work. You will build original problems involving simulations, result interpretation, experimental design, and extracting insights from data, then test and refine them to achieve an appropriate level of difficulty.

Key Responsibilities

• Create computational problems requiring skilled use of specialized scientific software.
• Develop fully defined, multi-step workflows where an AI must calculate exact answers correctly.
• Design open-ended problems that require strategic planning of queries or experiments, interpretation of partial results, and efficient narrowing of possible answers.
• Test each problem against advanced AI models and revise the design based on results.
• Write Python-based problem setups, oracle functions, and solution validators.
• Create tasks grounded in real computational chemistry research workflows, with difficulty driven by careful reasoning rather than raw computation.

Qualifications

• Graduate-level training in a relevant STEM field, including an MS, PhD, or equivalent research experience. An MS or PhD is preferred.
• Hands-on expertise in computational chemistry and electronic structure methods, including use of PySCF for Hartree-Fock, DFT, TDDFT, CASSCF, and post-HF calculations.
• Ability to develop problems involving excited-state analysis, orbital diagnostics, method selection for challenging electronic structures, and interpretation of artifacts caused by method limitations.
• Experience using scientific software libraries in real research, demonstrated through publications, open-source contributions, or professional work.
• Strong Python programming skills and practical understanding of tool limitations, edge cases, and research-level problem solving.
• Ability to work independently, incorporate feedback, and refine problem designs.
• Comfort working in a Linux terminal environment with remote compute sandboxes.
• Experience with other specialized software in computational chemistry or electronic structure will be considered.

Preferred Qualifications

• Experience across multiple relevant scientific domains or tools.
• Familiarity with benchmark or evaluation design.
• Background in scientific teaching or creating exams and problem sets.
• Experience with computational reproducibility and containerized environments.

Work Terms

• Remote, hourly engagement.
• Availability of at least 15 to 20 hours per week is required.

Compensation

• $70 to $100 per hour.