Applied Physics
Posted Today by Alignerr
Negotiable
Undetermined
Remote
Oxford, England, United Kingdom
Summary: The role of an Applied Physics AI Data Trainer involves leveraging deep expertise in physics to evaluate and enhance the reasoning capabilities of AI models. Candidates will develop complex physics problems, author benchmark solutions, and audit AI-generated outputs for physical consistency. This fully remote position requires a PhD in a relevant field and offers flexible hours ranging from 10 to 40 per week.
Key Responsibilities:
- Develop complex, open-ended physics problems at PhD qualifying exam level.
- Write rigorous, step-by-step "golden responses" as definitive benchmarks.
- Evaluate AI-generated simulations and proofs for physical consistency.
- Provide structured feedback to train models on physical reasoning.
- Document breakdowns in AI reasoning when faced with advanced physics challenges.
Key Skills:
- PhD completed or in final stages in Applied Physics, Physics, Engineering Physics, or a closely related field.
- Deep mastery of Classical Mechanics, Electrodynamics, Statistical Mechanics, and Quantum Mechanics.
- Exceptional analytical writing skills.
- Attention to detail in scientific notation and logical proof structure.
- Self-motivated and reliable in independent technical tasks.
- Experience with data annotation or scientific dataset evaluation is a plus.
- Proficiency in computational tools such as Python, MATLAB, or COMSOL.
- Background in research involving simulation, modeling, or experimental physics.
- Familiarity with AI tools or language model evaluation as an end user.
Salary (Rate): £32.00 hourly
City: Oxford
Country: United Kingdom
Working Arrangements: remote
IR35 Status: undetermined
Seniority Level: undetermined
Industry: IT
Applied Physics — AI Data Trainer
About The Role
What if your deep expertise in physics could directly shape how AI understands the physical world? We're looking for PhD-level Applied Physicists to stress-test cutting-edge Large Language Models — exposing the gaps in their reasoning and helping ensure they never violate the fundamental laws of the universe. This is a fully remote, flexible contract role. No prior AI experience needed — just mastery of physics and a rigorous, analytical mind.
Organization: Alignerr
Type: Hourly Contract
Location: Remote
Commitment: 10–40 hours/week
What You'll Do
- Design Advanced Problems: Develop complex, open-ended physics problems at PhD qualifying exam level — requiring multi-step reasoning, mathematical derivation, and deep conceptual understanding across quantum mechanics, electromagnetism, thermodynamics, and beyond
- Author Gold-Standard Solutions: Write rigorous, step-by-step "golden responses" that serve as the definitive benchmark — every unit, constant, and logical step accounted for
- Audit AI Reasoning: Evaluate AI-generated simulations, proofs, and explanations for physical consistency — identifying where models "hallucinate" physics that violates first principles
- Refine Model Behavior: Provide structured, expert feedback that trains models to reason correctly about boundary conditions, conservation laws, symmetry principles, and physical constraints
- Probe Failure Modes: Systematically document how and where AI reasoning breaks down when confronted with research-level physics challenges
Who You Are
- PhD completed or in final stages in Applied Physics, Physics, Engineering Physics, or a closely related field
- Deep mastery across the core pillars: Classical Mechanics, Electrodynamics, Statistical Mechanics, and Quantum Mechanics
- Exceptional analytical writing skills — you can explain a complex derivation with precision and clarity
- Uncompromising attention to detail: units, scientific notation, dimensional analysis, and logical proof structure are second nature to you
- Self-motivated and reliable when working independently on technical tasks
- No prior AI or data annotation experience required
Nice to Have
- Experience with data annotation, scientific dataset evaluation, or quality assurance workflows
- Proficiency with computational tools such as Python (NumPy/SciPy), MATLAB, or COMSOL
- Background in research involving simulation, modelling, or experimental physics
- Familiarity with AI tools or language model evaluation as an end user
Why Join Us
- Work on high-impact AI projects alongside world-leading research labs
- Fully remote and flexible — work when and where it suits you
- Freelance autonomy with the structure of meaningful, intellectually stimulating work
- Directly influence how AI understands and reasons about the physical world
- Potential for ongoing work and contract extension as new projects launch