Computational Genomics AI Benchmarking Expert

You will create realistic computational genomics and bioinformatics problems for evaluating AI agents. Each benchmark must require biological interpretation, multi-step computation, code execution, and results that can be checked objectively.

You will also build reproducible reference solutions, test scientific correctness and difficulty, and document the work clearly for review.

• Design tasks using sequence, annotation, expression, and variant data.
• Define inputs, expected output formats, deterministic ground truths, and grading criteria.
• Develop reference workflows with Python, command-line tools, and bioinformatics libraries.
• Test tasks for solvability, scientific accuracy, appropriate difficulty, and resistance to plausible but incorrect answers.
• Incorporate reviewer feedback and communicate technical requirements or blockers.

What it pays and takes

This is a remote contractor assignment lasting five weeks. It is listed as entry level, but the work requires substantial research and computational genomics experience.

• $260 per approved task.
• 20+ hours per week.
• Remote and open worldwide.
• Four hours of daily overlap with Pacific Time is required.
• Professional fluency in English.
• A Ph.D., postdoctoral experience, or equivalent research experience in bioinformatics, computational biology, genomics, computational genetics, or a related field.
• Strong hands-on Python experience and experience analyzing biological sequence or genomics datasets.
• Comfort with Linux and command-line computational environments.
• Ability to assess ambiguous data, identify artifacts, choose suitable analytical methods, and decide when results are ready for use.
• Helpful experience includes variant analysis, transcriptomics, sequence analysis, phylogenetics, population genetics, functional genomics, clinical genomics, or reproducible scientific pipelines.
• Useful tools and technologies include Biopython, pandas, NumPy, SciPy, samtools, bcftools, BLAST, PLINK, Docker, and containerized workflows.
• Experience evaluating AI systems, designing benchmark datasets, creating automated graders, or publishing computational genomics research is also useful.
• This contractor assignment does not include medical leave or paid leave.

How it works

Apply on OpenTrain with your resume, then complete the application on the hiring site.

About AI training work

AI training is the human work used to improve artificial intelligence, including preparing examples, writing evaluation tasks, and checking whether model outputs are correct. People with deep scientific or technical expertise are needed for specialized projects because they can create reliable examples and judge complex results.