Role Title: Data Science & Analytic Methodology SME Role Type: Contract Location: Remote Required Skills: • LLMs • Pipeline integration • Data analytics • Scaled-analysis judgment • Data science In this role, you'll apply your expertise to help train next-generation AI systems. Your work will shape how models learn, reason, and perform through high-quality, real-world input. No prior experience in AI is required - your domain knowledge is what matters. Scope of Work: • Map scaled-analysis and pipeline-integration scenarios to an established taxonomy across key evaluation areas. • Author detailed analytic cases that reflect authentic deployment patterns, including automated triage, enrichment, and analysis-at-scale, as well as representative failure modes in real-world settings. • Evaluate and grade responses based on criteria such as source handling, reasoning, understanding of limitations, and practical utility. • Identify and flag instances of model over-refusal that could hinder legitimate analytic workflows or disrupt pipeline continuity. • Document challenging edge cases, emerging methodology concerns, and provide actionable recommendations for continuous improvement. • Contribute clear, insightful written and verbal feedback that supports iterative, high-quality project outcomes. Preferred Qualifications: • Demonstrated experience integrating LLMs—such as government GPT variants—into complex data or analytic pipelines. • Proven capability in conducting analysis at scale, with an understanding of automation failure modes and relevant evaluation metrics. • Track record of teaching, mentoring, or technical leadership in data science or related analytic disciplines. • Expertise in authoring clear, concise technical documentation and conveying complex concepts through strong written English. • Exposure to the design of evaluation frameworks or benchmark systems. • Familiarity with defense, intelligence, or government analytic domains is highly valued. • Ability to operate effectively in distributed, international project environments with minimal clearance constraints. Originally posted on Himalayas