Computational Chemist for AI Model Evaluation

Design original, graduate-level computational chemistry problems that test whether advanced AI systems can perform research-grade electronic structure work using real scientific software. The role focuses on creating tasks that require running simulations, interpreting results, designing experiments, and extracting hidden information from data. This is not a data-labeling role, it requires domain expertise to build challenges that demand strategic scientific reasoning rather than rote computation.

• Create original problems that require skilled use of specialized scientific software, including fully specified setups that demand correct execution of multi-step workflows.
• Design harder, open-ended tasks where the AI must plan measurements or queries, interpret partial outputs, and iteratively narrow hypotheses.
• Author problem setups, oracle functions, and automated solution validators in Python.
• Test each problem against state-of-the-art AI models, evaluate performance, and refine problems until they meet target difficulty.
• Document expected results, common failure modes, and computational artifacts tied to method limitations.
• Work independently to improve problem designs based on testing outcomes and feedback.

• Primary domain: Computational Chemistry and Electronic Structure.
• Hands-on experience with PySCF for quantum chemistry calculations, including Hartree-Fock, density functional theory, time-dependent DFT, CASSCF, and post-Hartree-Fock methods.
• Problem topics may include excited-state analysis, orbital diagnostics, selecting appropriate methods for challenging electronic structures, and interpreting artifacts caused by method choices.
• Experience with other specialized software in this domain is also relevant and will be considered.

• Graduate-level training in a relevant STEM field, MS or PhD preferred, or equivalent research experience.
• Proven, hands-on proficiency with at least one relevant scientific software library, demonstrated via research publications, open-source contributions, or professional work.
• Strong Python programming skills, including experience writing problem setups and automated validators.
• Practical research experience using quantum chemistry tools, understanding edge cases and failure modes.
• Comfortable working in a Linux and terminal environment with remote compute sandboxes.
• Ability to work independently, iterate on designs, and respond to testing feedback.
• Available for at least 15 to 20 hours per week.

• Experience across multiple related domains or software tools.
• Familiarity with benchmark or evaluation design for scientific tasks.
• Background in scientific teaching, exam writing, or problem-set design.
• Experience with computational reproducibility and containerized environments.

Remote engagement, hourly contract. Expected minimum availability is 15 to 20 hours per week. Work involves using a Linux/terminal environment and interacting with remote compute sandboxes to develop and validate problems. Candidates will write code and validators in Python as part of the engagement.

Pay rate: 70 to 100 per hour.

• Must have the graduate-level training and demonstrable software proficiency described above.
• Must be able to work in a Linux/terminal environment and use remote compute resources.
• Must be able to commit to at least 15 to 20 hours per week for the engagement.