AI Training Specialist – Physics at Turing

AI Training Specialist – Physics | Contractor | Remote

Join Turing, the world's leading research accelerator for frontier AI labs, to design challenging physics problems that train and evaluate advanced AI models. This role combines deep physics expertise with cutting-edge AI development.

About Turing:

Based in San Francisco, Turing is a trusted partner for global enterprises and frontier AI labs. We accelerate AI research with high-quality data, advanced training pipelines, and top researchers specializing in coding, reasoning, STEM, and multimodality.

Role Overview:

Design rigorous upper-undergraduate and graduate-level physics problems to train frontier AI models. Create exam-quality tasks that challenge advanced language models while remaining solvable by human experts.

Main Responsibilities:

• Design upper-undergraduate and graduate-level physics problems with complete step-by-step solutions
• Create model-breaking tasks that resemble top university coursework but challenge frontier AI models
• Customize problem sets to strict specifications (SI units, LaTeX formatting, explicit constants, closed-form answers)
• Convert multi-part or image-based questions into single, self-contained, text-only problems
• Write clear, rigorous solution processes and verify consistency across all components
• Evaluate model outputs, diagnose failure modes, and iteratively refine problems
• Review and quality-check physics problems created by other trainers
• Collaborate with trainers, reviewers, and project managers to maintain quality at scale

Required Qualifications:

• Master's degree or PhD in Physics or closely related field
• Strong background in upper-undergraduate physics or beyond
• Fluent English with ability to write precise scientific text
• Strong LaTeX proficiency for mathematical derivations and formatting
• Experience with university teaching, grading, exam design, or analytical research
• Excellent attention to detail in units, constants, symbols, and notation
• Ability to verify and recompute multi-step quantitative solutions
• Familiarity with modern AI/LLM concepts (SFT, RLHF, evaluation)
• Expertise across physics domains: Classical Mechanics, Electromagnetism, Thermodynamics & Statistical Physics, Quantum Mechanics, Optics and Modern Physics
• Ability to perform multi-step derivations, numerical calculations, and physical plausibility checks

Nice-to-Have:

• Prior experience creating content for AI training or standardized testing
• Experience designing multiple-choice questions with strong distractors

Offer Details:

• Commitment: 40 hours per week with 4-hour overlap with PST
• Engagement: Contractor assignment (no medical/paid leave)
• Duration: 1 month with possibility to extend
• Location: India, Bangladesh, Egypt, Kenya, Mexico, Ghana, Nigeria, and Turkey

Perks:

• Fully remote work environment
• Work on cutting-edge AI projects with leading LLM companies
• Contract extension opportunities based on performance

Evaluation Process:

• 15-minute delivery interview