Computational Life Sciences Expert

Gramian Consultancy is a boutique consultancy specializing in IT professional services and engineering talent solutions. With a strong background in software engineering and leadership, we help companies build high-performing teams by matching them with professionals who truly fit their needs.

The Role

We are seeking Computational Life Sciences experts with strong scientific programming skills to support the development of advanced AI systems. In this role, you will design realistic, multi-step scientific workflows where AI agents must navigate scientific data, write and execute code, use computational tools, troubleshoot intermediate results, and produce scientifically valid outputs.

You will help evaluate whether AI systems can perform authentic computational scientific work—not simply answer scientific questions. Tasks must be reproducible, self-contained, and executable in controlled, network-isolated environments.

Key Responsibilities

• Design challenging, realistic agentic scientific workflows across the life sciences.
• Create multi-step computational tasks requiring scientific reasoning and execution.
• Develop realistic datasets, input files, instructions, constraints, and expected deliverables.
• Design workflows requiring agents to inspect data, select methods, execute analyses, troubleshoot issues, and synthesize results.
• Create reproducible expert solutions and objectively verifiable ground truths.
• Develop robust automated or semi-automated grading criteria.
• Ensure tasks evaluate scientific reasoning and execution rather than memorization.
• Validate scientific assumptions, calculations, code, intermediate outputs, and final results.
• Ensure tasks are self-contained and executable in controlled computational environments.
• Maintain high task quality and throughput while incorporating reviewer feedback.

Environment

Tasks are executed in controlled, computationally isolated environments. Depending on the workflow, projects may involve:

• Python and scientific programming libraries

• Command-line tools

• Domain-specific scientific software

• Scientific datasets and structured input files

• Packaged dependencies and reproducible computational resources

Experts must design tasks that can be reliably packaged, executed, and evaluated within the project environment.