Data Scientist / Quantitative Analyst for AI Model Evaluation

Help shape rigorous evaluation benchmarks for frontier AI models by turning real-world analytical work into challenging, reproducible tasks. You will partner closely with researchers to identify where models succeed or fall short in data cleaning, statistical analysis, interpretation, and reporting. Role Overview

Design complex analysis assignments that mirror research workflows, such as comparing anomaly-detection algorithms on a dataset, calculating correlations, conducting manual spot checks, and summarizing findings in a notebook that supports a research decision. Each assignment typically requires one to two days of continuous, focused work.

• Create realistic data-analysis challenges involving messy-data cleaning, method comparison, and result interpretation.
• Complete the tasks you design in Jupyter Notebooks or Google Colab, producing clear, reproducible reference analyses.
• Develop fair comparisons between analytical approaches, supported by spot checks and a clear recommendation.
• Evaluate model responses to determine whether their statistical work and conclusions are valid.
• Collaborate with researchers and fellow experts to maintain consistent, accurate evaluations.

• Master''s degree or PhD in statistics, data science, or another quantitative STEM field, or equivalent practical experience in a research-intensive analytical domain.
• At least 1 year of experience in research, research engineering, or a data-analysis-intensive role.
• Hands-on expertise in data cleaning, statistical correlation, hypothesis testing, and careful interpretation of results.
• Proficiency with Jupyter Notebooks or Google Colab.
• Working proficiency with Python, including pandas, NumPy, or similar tools, and Git.
• Strong written communication skills for explaining analytical findings to decision-makers.
• Experience with AI training, model evaluation, or benchmark or task authoring is preferred.
• Exceptional attention to detail, creativity in task design, and the ability to independently solve ambiguous, open-ended problems.

• Remote role available within the United States.
• Full-time W-2 employment, with placement on an AI lab''s extended workforce.
• Approximately 35 hours per week, with reliable ongoing availability required.
• The employer of record administers employment, payroll, benefits, and compliance.

$60 to $90 per hour.

Employment decisions are made without discrimination based on any legally protected characteristic. Reasonable accommodations are available for qualified individuals with disabilities and disabled veterans throughout the application process.