Senior Aerodynamics Expert - AI Benchmark Problem Writer

About the role

We're building a rigorous benchmark of the hardest free-response reasoning problems in aerodynamics and thermofluids engineering — questions so demanding that even today's most capable AI models fail to solve them correctly. We need a senior subject-matter expert to author original, industry-grounded problems, produce complete expert-level solutions, and rigorously validate that these problems genuinely stump frontier AI systems.

This is a unique opportunity to shape how the next generation of AI models is evaluated on deep technical reasoning, working at the intersection of aerospace engineering and AI research.

What you'll do

• Author original, free-response engineering problems in aerodynamics and thermofluids that reflect real-world industry scenarios and challenges
• Write complete, expert-level, step-by-step solutions for each problem you create
• Validate the difficulty of each problem by testing it against at least three frontier large language models
• Iterate on problem design until the question reliably causes at least one frontier model to fail, ensuring true benchmark-level difficulty
• Document reasoning, assumptions, and edge cases clearly so problems can be independently reviewed and graded
• Collaborate with a small team of technical reviewers to refine question quality, clarity, and correctness

Requirements

• PhD in Aerospace Engineering, or Mechanical Engineering with a specialization in thermofluids/aerodynamics
• 5+ years of professional or research experience in aerodynamics, fluid dynamics, or related engineering domains
• Deep, demonstrable expertise in solving advanced, real-world aerodynamics problems
• Strong technical writing skills — able to produce clear, rigorous, and pedagogically sound problem statements and solutions
• Familiarity with or willingness to learn how to interact with and evaluate outputs from large language models
• Meticulous attention to detail and comfort with an iterative, quality-driven review process
• Nice to have: prior experience creating technical assessments, exam questions, or academic benchmarks; familiarity with AI/ML evaluation methodologies

Compensation

This role is compensated at $85/hour, fully remote, and open globally to qualified experts.