Physics AI Trainer

MissionHires; United States; Remote; USD 55,000 - 172,000; Part time

Benefits offered by this job

Fully remote contract

Flexible schedule

Minimum 10 hours/week

30–40 hours typical per week

Pay via PayPal

Onboarding and expert review

Job summary

MissionHires is seeking experienced professionals to design and evaluate high-fidelity physics and engineering tasks for AI models. This fully remote contractor role emphasizes real-data workflows, rigorous assessment, and detailed rubrics to ensure safe and accurate model usage.

Ideal candidates have 2+ years in technical fields, strong Python or computing skills, and ability to operate without AI-assisted shortcuts.

Qualifications

• 2+ years of applied experience in physics, mathematics, chemistry, biology, engineering, data science, or related discipline.
• In-progress Bachelor's degree or higher; 2+ years of experience outside undergraduate study preferred.
• Current or recent hands-on individual-contributor experience in a technical field.
• Experience working with real technical data and field-specific tools (instrument data, measurement files, simulation outputs, schematics, drawings).
• Strong scientific and experimental reasoning skills, including experiment design, measurement error, uncertainty, and statistical pitfalls.
• Ability to assess work in adjacent sub-disciplines and explain what was done correctly or incorrectly.
• Strong written English communication skills and ability to explain why a technical result is incorrect.
• Familiarity with AI/LLM tools and ability to distinguish well-reasoned outputs from plausible but incorrect answers.
• Coding or scientific computing experience in Python preferred; comfort with command-line tools needed for some projects.
• Baseline proficiency with cloud file tools, browser profiles, desktop installation, and file-management tasks.
• Advanced degree or licensure is a plus; practical applied experience valued over credentials.
• Must complete the skills assessment independently without using AI tools.

Responsibilities

• Design realistic, challenging physics and engineering tasks based on professional workflows, including scenarios, prompts, and supporting files such as test data and schematics.
• Run tasks through frontier AI models and evaluate calculations, analyses, design reviews, and technical reports against professional standards.
• Compare model outputs generated from identical prompts and files, determine which performed better, and document strengths and shortcomings.
• Write detailed grading rubrics defining assumptions, methods, magnitudes, units, failure modes, and deliverables for correct responses.
• Identify and document concrete model failures, including unit or scaling errors, misread data, unsupported conclusions, or missing safety considerations.
• Build professional-style task files ranging from compact multi-file scenarios to larger assignments spanning days.
• Review and refine tasks created by other experts across physics and related disciplines.
• Complete a proposed task based on professional experience, including scenario, prompt, source files, anticipated failure points, and full task build after approval.

Skills

Applied physics

Mathematics

Engineering

Data analysis

Python

Experiment design

Statistical reasoning

English writing

LLM tools

CLI familiarity

Cloud tools

Education

Bachelor's degree (in-progress)

2+ years outside undergrad

Tools

Python

CLI

Excel

Job description

When AI gets the details wrong in your field, real people pay for it. The people who know the work best have to be the ones training these models. You’re not being replaced by AI. You’re making sure it’s used accurately and safely.

Responsibilities

• Design realistic, challenging physics and engineering tasks based on professional workflows, including scenarios, colleague-style prompts, and supporting files such as test data, specifications, simulation outputs, lab records, schematics, drawings, and reports.
• Run tasks through frontier AI models and evaluate calculations, analyses, design reviews, and technical reports against professional standards.
• Compare model outputs generated from identical prompts and files, determine which performed better, and document strengths and shortcomings.
• Write detailed grading rubrics defining the assumptions, methods, magnitudes, units, failure modes, and deliverables required for a correct response.
• Identify and document concrete model failures, including unit or scaling errors, misread data, unsupported conclusions, fabricated or ignored source material, missed safety or boundary conditions, and off-brief interpretations.
• Build professional-style task files yourself, ranging from compact multi-file scenarios to larger assignments that may take several days to develop.
• Review, refine, and contribute to tasks created by other subject-matter experts across physics and adjacent disciplines.
• Complete a proposed task based on professional experience, including the scenario, prompt, source files, anticipated model failure points, and full task build after proposal approval.

Requirements

• 2+ years of applied experience preferred in physics, mathematics, chemistry, biology, engineering, data science, or a closely related technical discipline.
• In-progress Bachelor's degree or higher; 2+ years of experience outside undergraduate study is preferred.
• Current or recent hands‑on individual‑contributor experience in a technical field, rather than solely managerial experience.
• Experience working with real technical data and field‑specific tools, such as instrument or test data, measurement files, simulation outputs, schematics, drawings, and analysis software.
• Strong scientific and experimental reasoning skills, including experiment design, measurement error, uncertainty, and common statistical pitfalls.
• Ability to assess work in adjacent sub‑disciplines and explain what was done correctly or incorrectly.
• Strong written English communication skills and the ability to clearly explain why a technical result is incorrect.
• General familiarity with AI and LLM tools and the ability to distinguish well‑reasoned outputs from plausible but incorrect answers.
• Coding or scientific computing experience in Python or a similar language is strongly preferred and required for some projects; comfort with command‑line tools is also needed for coding‑based work.
• Baseline technical proficiency with cloud file tools, browser profiles, desktop application installation, and common file‑management tasks.
• Advanced degree or professional licensure is a plus but not required; practical applied experience is valued over credentials.
• Must be able to complete the skills assessment independently without using AI tools.

Benefits

• Pay of $40-$125 per hour USD, paid via PayPal on a regular cadence.
• Fully remote contract work with flexible scheduling.
• Minimum commitment of 10 hours per week, with no weekly maximum; many projects prioritize 30-40 hours per week and full‑time hours may be available.
• Multi‑day task timers that allow work to be spread across several days.
• Project access to Claude and ChatGPT; no personal subscription is required.
• Structured onboarding, training, expert review, written feedback, and opportunities to review other experts’ work.
• Compensation for each completed stage of the project process, subject to acceptance requirements.