Bioinformatics Computational Genomics Expert

You will create difficult, realistic computational genomics and bioinformatics problems used to evaluate advanced AI systems. Each task should test whether an AI agent can interpret biological data, write and run code, use scientific file formats, and produce results that can be checked objectively.

You will combine scientific judgment, computational analysis, benchmark design, reference-solution development, and detailed quality review. You will also document your work and communicate requirements, blockers, and progress to project leads and reviewers.

• Design tasks using FASTA, FASTQ, VCF, BAM, SAM, BED, TSV, and CSV files.
• Work with sequence annotations, expression data, and germline or somatic variant data.
• Create task specifications, input datasets, expected output formats, deterministic ground truths, and expert reference solutions.
• Build reproducible workflows with Python, Linux command-line tools, and established bioinformatics libraries.
• Check scientific correctness, solvability, reproducibility, and difficulty for advanced AI models.
• Create grading criteria that separate scientifically correct solutions from plausible-looking but incorrect results.

This is a remote contractor assignment lasting five weeks. The work requires substantial daily and weekly availability, including overlap with Pacific Time.

• $150 per approved task.
• At least 6 hours per day and 40 hours per week.
• At least 4 hours of overlap with Pacific Time.
• Remote work with candidates accepted worldwide.
• Ph.D., postdoctoral experience, or equivalent research experience in bioinformatics, computational biology, genomics, computational genetics, or a related field.
• Strong hands-on Python programming experience.
• Experience analyzing biological sequence or genomics datasets.
• Comfort working in Linux and command-line computational environments.
• Strong understanding of the biological and experimental context behind computational analyses.
• This assignment does not include medical or paid leave.

Experience in variant analysis, transcriptomics, sequence analysis, phylogenetics, population genetics, functional genomics, or clinical genomics is useful. Familiarity with common scientific tools and experience evaluating AI systems can also help.

• Biopython, pandas, NumPy, SciPy, samtools, bcftools, BLAST, PLINK, or equivalent tools.
• AI or large language model evaluation.
• Benchmark dataset or automated grader design.
• Reproducible pipeline development.
• Docker and containerized scientific workflows.

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

AI training is the human work behind artificial intelligence, including creating examples, testing model responses, and checking whether results are accurate. People with deep scientific or technical knowledge are needed to create reliable evaluation tasks that measure what AI systems can actually do.