• Identify high-leverage DSD use cases: Target opportunities across DS operations (technical knowledge retrieval, investigation support, documentation acceleration, operational insights, exception triage, capacity/coordination decision support). • Build prototypes and deliver MVPs: Create working solutions using AWS/Databricks/Dataiku/OpenAI, integrating operational data and documents into usable workflows. • Data readiness and integration: Assess data availability/quality; define what's needed and influence how to get there (standard sources, access patterns, reusable pipelines). • Roadmap ownership: Produce a DSD capability roadmap with milestones, dependencies, and impact metrics. • Balance quick wins vs foundations: Deliver what helps DSD now while shaping patterns that will scale as the enterprise platform matures (reusable assets, integration standards, evaluation/monitoring, agent orchestration approaches). • Vision and transition plan: Define the DSD AI north star and a staged path from POCs to reusable capabilities to scaled delivery. • Drive ChatGPT adoption: Build prompt libraries, playbooks, and training tied to real DS work; measure usage and outcomes. • Stay current and apply responsibly: Keep pace with AI advances and translate them into practical, controlled, high-impact solutions suitable for DSD environments.