The work
You will create expert-level training data for large language models, which are AI systems that generate text. The work uses research knowledge to develop prompts, judge responses, and explain complex STEM ideas clearly.
The role title and the duties and qualifications should be reviewed before applying.
- Develop prompts based on research knowledge in a STEM field.
- Evaluate AI responses for accuracy, relevance, and quality of explanation.
- Research technical topics with help from AI.
- Turn complex concepts into clear task instructions and evaluation criteria.
- Create training data using careful scientific and technical judgment.
What it pays and takes
This is a remote, asynchronous contractor role. Projects operate year-round and open periodically based on current project needs.
- Pay: Up to $48 per hour.
- Time: 20+ hours per week.
- Location: Open to candidates in the United States.
- Language: Professional fluency in English.
- Education and experience: Doctoral study, a PhD, or postdoctoral experience in mathematics, chemistry, biology, physics, or engineering.
- Skills: Strong written communication and the ability to explain complex concepts clearly.
- Skills: Ability to assess whether AI systems understand and explain important concepts in your field.
- Work style: Able to work independently in a remote, asynchronous setting.
- Prior AI experience is not required.
How it works
Apply on OpenTrain with your resume, then complete the application on the hiring site.