Life Sciences Scientific Reasoning Task Designer

You will create advanced scientific reasoning tasks for AI evaluation. The work uses realistic research scenarios and asks you to turn biological evidence into clear, rigorous questions and reference answers.

Your work will focus on scientific interpretation and reasoning rather than programming or code execution. You will also review task quality, address ambiguity, and incorporate reviewer feedback.

• Design novel, graduate-level life sciences problems based on realistic research contexts.
• Write precise, self-contained questions and defensible reference answers.
• Interpret experiments, datasets, figures, sequences, structures, and other biological evidence.
• Validate biological claims, calculations, assumptions, and conclusions.
• Identify alternative interpretations and questions that are not sufficiently constrained.
• Document tasks clearly and make them reproducible where appropriate.
• Maintain quality and throughput while responding to reviewer feedback.

This is a remote contractor assignment for one month. The role description calls for at least 40 hours per week, including four hours of overlap with Pacific Time and at least eight hours per day; the listing also records a 20+ hour weekly commitment.

Compensation is described as competitive, but no specific rate or payment amount is provided. The role is listed in USD and may be suitable for someone seeking part-time contract work.

• Engagement:
  One-month remote contractor assignment.
• Schedule:
  
  At least 40 hours per week in the role description, with four hours of Pacific Time overlap and a minimum of eight hours per day.
• Listed time requirement: 20+ hours per week.
• Language:
  English.
• Education or experience:
  Ph.D., postdoctoral experience, or equivalent advanced research experience in the life sciences.
• Expertise:
  Strong knowledge of at least one biological discipline.
• Skills:
  
  Scientific experiment and data interpretation, scientific writing, precision, and the ability to explain complex concepts clearly.
• Helpful background:
  Graduate-level problem or assessment design, manuscript or research review, AI-generated scientific response evaluation, or AI and machine learning evaluation datasets.
• Relevant fields include molecular biology, genetics and genomics, cell biology, biochemistry, neuroscience, microbiology, immunology, evolutionary biology, ecology, structural biology, biophysics, developmental biology, cancer biology, pharmacology, and related life sciences.
• Useful additions include peer-reviewed publications or graduate-level teaching and mentoring experience.

AI training work uses human-written examples, evaluations, and feedback to help improve how AI systems answer questions and reason. Open Train is a platform for finding and building careers in this field, where contributors can create a profile and apply for projects in one place.