Create graduate-level life sciences problems that test how AI interprets experiments, biological data, figures, sequences, structures, and research evidence. This remote contractor role requires advanced life sciences research experience.
The work
You will design advanced scientific reasoning tasks for AI evaluation. The work focuses on biological knowledge, experimental interpretation, data analysis, and careful scientific reasoning rather than coding.
You will turn realistic research scenarios and evidence into challenging problems that test synthesis and interpretation instead of simple factual recall. You will also write clear questions and defensible reference answers.
- Design graduate-level life sciences problems.
- Interpret experiments, datasets, figures, sequences, structures, and literature-style evidence.
- Validate biological claims, calculations, assumptions, and conclusions.
- Find ambiguity and alternative interpretations in scientific tasks.
- Make sure each task is self-contained and clearly documented.
- Use reviewer feedback while maintaining quality and throughput.
What it pays and takes
The listing does not provide a pay rate. This is a remote contractor assignment focused on life sciences scientific reasoning and written evaluation work.
- Pay: Not specified in the listing.
- Engagement: Contractor, listed as part-time.
- Assignment length: One month.
- Schedule: At least 40 hours per week and at least 8 hours per day.
- Time-zone overlap: At least 4 hours overlapping with Pacific Time.
- Location: Remote and open worldwide.
- Language: English.
- Education or experience: Ph.D., postdoctoral experience, or equivalent advanced research experience in the life sciences.
- Expertise: Strong knowledge in at least one biological discipline.
- Required abilities: Interpret scientific experiments, data, and research findings; write clearly and precisely; and pay close attention to detail.
- Helpful experience: Designing graduate-level problems, reviewing scientific research, or evaluating AI-generated scientific responses.
- The structured listing notes 20+ hours per week, while the role description requires at least 40 hours per week.
How it works
Apply on OpenTrain with your resume, then complete the application on the hiring site.
About AI training work
AI training work is the human side of building AI: people create examples, evaluate model responses, and prepare reliable data. Experienced researchers are needed when tasks require careful interpretation, domain knowledge, and defensible scientific answers.