Your deep understanding of physical principles will directly shape how AI reasons through complex scientific problems — making a real impact on the future of intelligent systems.
This is a fully remote contract role designed to fit around your existing schedule. Whether you're a graduate student, researcher, or working physicist, you can contribute on your own terms.
Why Join Us
Work on frontier AI projects alongside leading research labs.
Fully remote and flexible — work when and where it suits you.
High autonomy and contractor independence.
Contribute to meaningful work that advances AI's scientific reasoning capabilities.
Potential for ongoing contract renewals and expanded project involvement.
Responsibilities
Design challenging, rigorous physics problems spanning mechanics, electromagnetism, thermodynamics, quantum mechanics, and more.
Develop clear, step-by-step solutions with well-structured logical reasoning.
Evaluate AI-generated responses for scientific accuracy, reasoning quality, and completeness.
Identify where AI models fall short and provide actionable, structured feedback.
Collaborate with researchers to build and refine benchmarks from undergraduate through Masters level.
Work independently and asynchronously on your own schedule.
Required skills
Strong problem-solving ability across multiple advanced physics domains.
Able to communicate complex ideas clearly and precisely in writing.
Detail-oriented with a systematic approach to evaluating quality.
Self-motivated and comfortable working independently.
Preferred skills
Experience in academic research, tutoring, or scientific writing.
Familiarity with data annotation, evaluation systems, or quality assurance workflows.
Exposure to machine learning or AI research environments.
Experience and education
Currently pursuing or have completed a Masters degree in Physics, Applied Physics, or a closely related field.
Thriveth makes AI data-training work easier to find, understand, and navigate. We replace uncertainty with clear opportunities, realistic expectations, and insights from real application journeys.