Materials Science AI Evaluation Task Author

About OpenTrain

OpenTrain is the #1 platform for finding and building careers in AI training and data labeling. It helps specialists discover projects, build a professional AI training profile, and apply in minutes. Creating an OpenTrain account is free.

As an OpenTrain contractor, you will contribute to advanced AI development while building a durable portfolio of scientific evaluation work.

• Worldwide remote opportunity
• Contractor and part-time engagement
• 20+ hours per week
• Rate: $70 USD per hour
• Working language: English

About AI Training and Scientific Evaluation

AI training is the human side of building artificial intelligence. People create examples, review model outputs, and define evaluation standards that help advanced systems reason more accurately and reliably.

In this role, your scientific expertise will help test whether AI models can solve authentic computational problems in materials science. Your work can directly shape how frontier models handle research-grade scientific reasoning and code.

• Work on cutting-edge AI evaluation
• Apply research expertise to executable scientific problems
• Help measure the reliability of advanced AI models

The Role

OpenTrain is recruiting a Materials Science AI Evaluation Task Author to create challenging scientific-computing tasks for evaluation of advanced AI models. The work focuses on semiconductor materials and molecular modeling and combines scientific judgment, programming, prompt design, grading-rubric development, and task-difficulty calibration.

You will turn authentic research material and original scenarios into executable problems. Tasks should be released only when strong models fail more often than they succeed, ensuring that each evaluation meaningfully tests model capability.

• Subject area: Scientific Computing AI Evaluation
• Core focus: Semiconductor materials and molecular modeling
• Role classification: Entry level
• Engagement: Part-time contractor

What You’ll Do

You will develop scientifically rigorous evaluation tasks from published research, public datasets, open-source repositories, or scenarios you design yourself. Each task must include a clear problem, executable code or workflow, and grading criteria that define a correct answer.

You will also calibrate tasks against frontier models and collaborate through GitHub pull requests with automated quality checks.

• Source suitable material from published papers, Kaggle datasets, and open-source repositories
• Design original scientific scenarios when appropriate
• Write original prompts grounded in selected research material
• Create grading criteria for correct and incorrect answers
• Use executable code and scientific reasoning in model evaluations
• Calibrate task difficulty against frontier models
• Release tasks only after strong models fail more often than they succeed
• Work through GitHub pull requests and automated quality checks

Required Qualifications

This role requires PhD-level research training and substantial subject-matter expertise. The listing is categorized as entry level, but applicants must meet the advanced scientific, programming, and research-workflow requirements below.

• PhD in materials science, materials engineering, applied physics, chemistry, chemical engineering, or a closely related field
• Demonstrated depth in semiconductor materials and molecular modeling
• Working proficiency in Python, R, or another relevant scientific-computing language
• Experience with Git or GitHub, Docker, and executable research workflows
• Ability to design rigorous scientific problems
• Ability to define grading criteria and evaluate AI-generated solutions
• Ability to apply rigorous scientific judgment when assessing model performance

Helpful Background

The following experience is helpful but not listed as a requirement. It can strengthen your ability to create realistic, technically demanding evaluation tasks.

• Publications in peer-reviewed journals
• Prior scientific software experience
• Research engineering experience

Why Work in AI Training

AI training and data-labeling work is a fast-growing part of the technology industry. Specialists contribute to systems that learn from carefully prepared examples, expert reviews, and rigorous evaluations.

The work is remote and flexible, allowing contributors to fit projects around other commitments while applying valuable professional expertise to the development of state-of-the-art AI.

• Remote work from anywhere with an internet connection
• Flexible part-time scheduling
• Direct influence on how advanced AI systems perform
• Opportunity to build experience in a rapidly growing field

How to Apply Through OpenTrain

Create a free OpenTrain account, build your profile around your materials science and scientific-computing experience, and apply in minutes. Highlight your semiconductor materials knowledge, molecular modeling background, programming proficiency, and experience with executable research workflows.

• Apply as a worldwide contractor
• Plan for 20+ hours per week
• Work in English
• Rate: $70 USD per hour