Build data and semantic models, documentation, context, and evaluations that improve conversational analytics powered by LLMs. Embed with stakeholder teams to understand analyst workflows and business needs, then develop self-service tooling and standardized analytic frameworks. Train non-data stakeholders on effective data and LLM usage, create reusable training materials, maintain evaluation processes, and track conversational analytics accuracy. Requires strong SQL, dbt modeling, BI, communication, teaching, problem-solving, and project ownership skills.