Remote STEM Researcher - AI Evaluation & Benchmark Design

Weekday 1; United States; Remote; USD 83,000 - 124,000; Full time

Job summary

Weekday 1 is seeking researchers to develop rigorous evaluation benchmarks for frontier AI models. The role focuses on transforming experimental design and data-driven analysis into sophisticated, multi-step benchmarks that challenge current AI systems.

You will work with AI researchers to uncover subtle reasoning errors and methodological flaws while contributing to the continuous refinement of evaluation methodologies. This is a fully remote, full-time engagement with about 35 hours per week.

Qualifications

• Master's degree or PhD in a STEM field or equivalent research experience.
• At least 1 year of experience in an active research role.
• Experience with Python, data analysis, simulation, ML, or scientific computing.
• Strong understanding of experimental design and statistical analysis.
• Familiarity with Git, IDEs, and notebook platforms (Jupyter or Colab).

Responsibilities

• Design complex benchmark tasks informed by real-world scientific workflows.
• Develop reference solutions using Python, notebooks, and computational tools.
• Define evaluation standards to distinguish sound reasoning from flawed conclusions.
• Review AI-generated solutions and identify methodological weaknesses.
• Collaborate with researchers to improve benchmark quality and rigor.
• Refine evaluation methodologies for advanced AI systems.

Skills

Python

Data analysis

Machine learning

Statistical analysis

Scientific computing

Education

Master's degree

PhD

Tools

Git

Jupyter

Google Colab

Job description

Weekday 1 is seeking researchers to develop rigorous evaluation benchmarks for frontier AI models. The role focuses on transforming experimental design and data-driven analysis into sophisticated, multi-step benchmarks that challenge current AI systems.

You will work with AI researchers to uncover subtle reasoning errors and methodological flaws while contributing to the continuous refinement of evaluation methodologies. This is a fully remote, full-time engagement with about 35 hours per week.