- Experience
- Any
- Salary
- USD 70 – USD 100 / hour
- Openings
- 1
- Posted
- 1 day ago
Job description
About the Role
Mercor brings together top-tier technical and creative specialists to support leading AI research labs. The company is based in San Francisco and is backed by investors such as Benchmark, General Catalyst, Peter Thiel, Adam D'Angelo, Larry Summers, and Jack Dorsey.
This contract position is focused on STEM computational scientific software and evaluation design, specifically in pharmacokinetics and systems biology. The work is remote and centered on building rigorous scientific tasks that help assess and improve AI systems.
Compensation and Work Arrangement
Pay is set at $70–$100 per hour. The expected commitment is 15–20 hours each week, and the role is fully remote.
Key Responsibilities
- Create advanced computational challenges using scientific software tools such as libRoadRunner, Tellurium, or SBML-based libraries.
- Build problems that evaluate whether AI systems can carry out multi-step scientific workflows and apply structured reasoning.
- Tune task difficulty by comparing results with leading AI models and adjusting the design based on calibration feedback.
- Partner with AI research teams to strengthen output quality and improve training data.
- Work autonomously and asynchronously while meeting deadlines and contributing to model performance improvements.
Required Qualifications
- Graduate-level preparation in a STEM field, such as an MS, PhD, or comparable research experience.
- Hands-on experience with at least one of the listed scientific software tools, supported by publications, open-source contributions, or professional usage.
- Strong Python skills for creating problem statements, oracle logic, and solution validators.
- Ability to work independently and refine task designs using feedback from calibration cycles.
- Comfort using Linux or terminal-based environments, including remote compute sandboxes.
Preferred Background
- Exposure to several of the referenced scientific domains or software tools.
- Prior experience with benchmark creation or evaluation design.
- Familiarity with scientific teaching methods or assessment design.
- Experience with reproducible research workflows and containerized setups.
Application Process
The application is expected to take about 20–30 minutes and includes three steps: upload your resume, complete an AI interview based on your resume, and submit the form.
Support
For more information about the interview flow and platform details, refer to the company’s documentation. If you need assistance, contact the support team by email.
Additional Information
The hiring team reviews applications daily. To be considered, applicants should finish both the AI interview and the remaining application steps.