The Work
The duties and qualifications below focus on creating life sciences scientific reasoning tasks for AI evaluation. Review them carefully before applying.
You will turn realistic biological research scenarios, experimental findings, datasets, figures, sequences, structures, and literature-style evidence into challenging problems with objectively defensible answers. The work focuses on scientific interpretation and judgment rather than programming or simple factual recall.
- Design original life sciences problems that require advanced scientific reasoning.
- Write precise questions and rigorous reference answers based on realistic research contexts.
- Check biological claims, calculations, assumptions, and conclusions for accuracy.
- Identify ambiguity, alternative valid interpretations, and questions that are not sufficiently constrained.
- Document work clearly and reproducibly, use reviewer feedback, and maintain quality.
- Create tasks that distinguish AI models by how well they reason about scientific evidence.
What It Pays and Takes
The listing does not provide a pay rate. This is a remote contract opportunity for an individual scientific contributor, with part-time work of 20 or more hours per week.
- Open worldwide.
- Work 20 or more hours per week.
- Part-time contract role.
- Fluent in English.
- A Ph.D., postdoctoral experience, or equivalent advanced research experience in the life sciences is required.
- Strong expertise in at least one biological field, such as molecular biology, genetics and genomics, cell biology, biochemistry, neuroscience, microbiology, immunology, ecology, structural biology, biophysics, developmental biology, cancer biology, or pharmacology.
- Ability to interpret scientific experiments, biological data, and research findings.
- Ability to turn complex scientific ideas into precise, self-contained problems with defensible answers.
- Programming and code execution are not required.
- Helpful experience includes graduate-level assessment design, research or manuscript review, evaluation of AI-generated scientific responses, AI evaluation datasets, peer-reviewed publications, or graduate teaching and mentoring.
How It Works
Apply on OpenTrain with your resume and then complete the application on the hiring site.
About AI Training Work
AI training is the human work behind artificial intelligence, including writing and evaluating model responses, preparing examples, and checking whether systems reason accurately. People with deep subject knowledge help create reliable training and evaluation material for specialized topics such as the life sciences.