Materials Scientist for AI Model Evaluation

Help advance frontier AI systems by bringing rigorous materials science and engineering judgment to the evaluation, design, and improvement of technical knowledge work. You will work closely with an AI research team to define what high quality materials reasoning looks like in practice and ensure model outputs can withstand technical scrutiny.

• Review materials science tasks and model outputs, identifying missing behaviors, weak reasoning, unsupported structure to property claims, and technically unsubstantiated conclusions.
• Write instruction specifications and golden solutions for materials problems, and create tasks that reflect real materials science and engineering work.
• Design challenging evaluation sets and benchmarks that measure progress in materials-specific reasoning.
• Partner with researchers and adjacent domain specialists to develop materials-focused skills and tools.
• Translate expert, tacit judgment into clear, teachable criteria and maintain consistent evaluation standards.

• PhD in materials science, materials engineering, or a closely related field such as chemistry, chemical engineering, applied physics, or metallurgy. A master''s degree with exceptional industrial depth may be considered.
• At least 4 years of substantive materials research or industrial R&D experience at a research university, national laboratory, or industrial research organization. Graduate coursework alone does not qualify.
• Specialization in at least one area, such as energy storage and battery materials, semiconductors and electronic materials, polymers and soft matter, structural alloys and metallurgy, characterization and microscopy, or computational materials and simulation.
• Senior-level research ownership, demonstrated through roles such as Senior Scientist, Staff Scientist, Research Lead, Principal Investigator, or a comparable senior industrial R&D position.
• Peer-reviewed publications, granted patents, or delivered materials programs are strongly preferred.
• Hands-on professional use of large language models and the ability to distinguish sound technical reasoning from plausible but incorrect answers.
• Excellent written communication and the ability to provide precise, well-structured feedback.

• Full-time W-2 employment, working 40 hours per week.
• Hybrid position based in the Bay Area, California, with on-site collaboration multiple days per week as required.
• Client-issued accounts and equipment will be provided, and work will be performed within the client’s tools alongside its research teams.

• $70 to $110 per hour.

• You must live in the Bay Area or relocate there at your own expense before the engagement begins. Relocation assistance is not available.
• Qualified applicants receive equal employment opportunity without discrimination based on legally protected characteristics. Reasonable accommodations are available for qualified individuals with disabilities and disabled veterans during the application process.

• Submit your application through the opportunity channel. If selected, employment, onboarding, payroll, benefits, and compliance are managed by the employer of record while you work as part of the client team.