Agentic Life Science Expert

We are looking for experts in Computational Life Sciences with strong scientific programming skills to develop complex, realistic tasks for Agentic Life Sciences. In this role, you will design workflows where AI agents independently navigate scientific data, write and execute code, utilize computational tools, troubleshoot intermediate results, and produce scientifically valid final outputs. The primary objective is to assess whether AI systems can perform authentic multi-step scientific work rather than simply answer scientific questions.

Key Responsibilities:
- Design challenging and realistic agentic scientific workflows across various life sciences disciplines.
- Create tasks that require multiple computational steps instead of simple scripts or straightforward question-answer formats.
- Develop realistic input files, scientific datasets, instructions, constraints, and clearly defined expected deliverables.
- Build tasks that require agents to inspect data, select appropriate analytical methods, execute analyses, troubleshoot issues, and synthesize results.
- Produce reproducible expert solutions and objectively verifiable ground truths that operate within a controlled computational sandbox.
- Design robust automated or semi-automated grading criteria to evaluate task performance.
- Ensure tasks emphasize scientific reasoning and execution over memorization.
- Validate scientific assumptions, calculations, code, intermediate outputs, and final answers for accuracy and reliability.
- Guarantee that all tasks are self-contained and executable in controlled, network-isolated computational environments.
- Maintain high quality and throughput while effectively incorporating reviewer feedback.

Required Qualifications:
- Ph.D., postdoctoral experience, or equivalent research background in life sciences with strong computational and scientific programming expertise.
- Proficiency in scientific programming, particularly using Python.
- Experience conducting multi-step computational scientific analyses.
- Ability to independently validate both scientific reasoning and computational outputs to ensure correctness.

Preferred Qualifications and Benefits:
- Expertise in one or more specialized areas such as bioinformatics, computational genomics, systems biology, computational neuroscience, biostatistics, computational drug discovery, computational biochemistry, structural biology, protein engineering, or computational microbiology.
- Familiarity with scientific libraries and command-line tools commonly used in computational research.
- Experience creating reproducible research pipelines that enhance reliability and transparency.
- Ability to translate authentic research workflows into bounded, objectively gradable tasks suitable for AI evaluation.
- Experience working with AI agents, coding agents, or scientific AI systems is a plus.
- Background in building automated evaluation environments is advantageous.
- Knowledge of Docker and Linux environments is beneficial.
- Publications in computational or data-intensive life sciences research will be considered an asset.

This position offers the opportunity to contribute to cutting-edge research by bridging computational life sciences and artificial intelligence, helping to push the boundaries of what AI can achieve in scientific discovery.