About this role
Role Overview
Design original, graduate-level computational science problems that evaluate whether advanced AI systems can complete research-level workflows using real scientific software. This work centers on pharmacokinetics and systems biology, including simulations, result interpretation, experiment design, and extracting hidden information from data.
Key Responsibilities- Create challenging problems requiring skilled use of specialized scientific software.
- Develop fully defined, multi-step tasks that require exact computational answers.
- Design open-ended problems in which an AI must strategically plan queries or experiments, interpret partial results, and efficiently narrow possible answers.
- Test problems against state-of-the-art AI models and refine them to reach the intended difficulty level.
- Write Python-based problem setups, oracle functions, and solution validators.
- Create problems that reward careful reasoning rather than surface-level pattern matching or raw computation alone.
- Graduate-level training in a relevant STEM field, including an MS, PhD, or equivalent research experience. An MS or PhD is preferred.
- Hands-on experience with pharmacokinetics and systems biology software, including libRoadRunner, Tellurium, SBML-based tools, or other specialized software in the domain.
- Experience with compartmental PK/PD modeling, enzyme kinetics, or systems biology simulations.
- Demonstrated proficiency with at least one listed scientific software library through research publications, open-source contributions, or professional work.
- Strong Python programming skills and experience applying these libraries to real research problems.
- Understanding of tool limitations, edge cases, and how to construct genuinely difficult scientific problems.
- Ability to work independently, incorporate feedback, and improve problem designs.
- Comfort working in a Linux terminal environment with remote compute sandboxes.
- Experience with multiple relevant tools or domains.
- Familiarity with benchmark or evaluation design.
- Experience teaching scientific subjects or creating exams and problem sets.
- Experience with computational reproducibility and containerized environments.
- Remote, hourly engagement.
- Availability of at least 15 to 20 hours per week is required.
$70 to $85 per hour.