Data Science Methodology SME

You will support AI training projects that test how well large language models handle real-world data science and analytic workflows. Your work will cover scaled analysis, pipeline integration, automation failures, over-refusal, and the practical value of model outputs.

You will turn analytic scenarios into structured evaluation cases, assess model responses, and explain where reasoning, source handling, limitations, or methodology need improvement.

• Map scaled-analysis and pipeline-integration scenarios to an established evaluation taxonomy.
• Write detailed cases covering automated triage, data enrichment, analysis at scale, and realistic failure modes.
• Evaluate and grade model responses for source handling, reasoning, limitations, and practical usefulness.
• Identify over-refusal that could block legitimate analysis or interrupt pipeline continuity.
• Document difficult edge cases and emerging concerns in analytic methodology.
• Give clear written and verbal recommendations to support iterative improvement.

This is a remote, part-time contractor role for candidates in the countries specified for the project. The work requires strong written English and the ability to explain complex technical ideas clearly.

• Pay: $60 to $100 per hour.
• Work type: Remote, part-time contract work.
• Location: Open to candidates in the listed countries.
• Language: Strong written English.
• Experience level: Entry level.
• Experience integrating large language models, including government GPT variants or equivalent systems, into complex data or analytic pipelines.
• Experience conducting analysis at scale, rather than only one-off analysis.
• Understanding of automation failure modes, evaluation metrics, and analytic methodology.
• Experience teaching, mentoring, or providing technical leadership in data science or a related field.
• Familiarity with evaluation frameworks or benchmark design.
• Helpful background: Defense, intelligence, or government analytics experience, plus the ability to work in distributed international environments with minimal clearance constraints.

You apply here in about two minutes and finish on the hiring platform. The next step is a short AI interview (technical roles include a coding challenge). Passing it is how you qualify for paid tasks, so take it soon after you are invited.

AI training is the human work behind modern artificial intelligence, including testing model answers, rating performance, and preparing examples that help systems improve. OpenTrain helps people find these projects, build a profile, and develop a portfolio of AI training and data-labeling experience.