# Roles and Responsibilities in an AI Team As organizations adopt [[Artificial Intelligence (AI)]], new roles emerge and existing roles evolve. Understanding who does what is critical for [[AI Implementation Roadmap|successful AI adoption]] and [[AI Governance]]. ## Core Roles ### Technical - **AI/ML Engineer**: builds, trains, deploys models; owns the [[AI Engineering]] stack including [[Retrieval-Augmented Generation (RAG)|RAG]] pipelines, [[AI Evaluation]], and [[AI Observability]] - **Prompt/Context Engineer**: designs [[Prompt Engineering|prompts]], [[AI Agent Skills|skills]], and [[Context Engineering]] systems; maintains [[Context-as-Code]] artifacts - **Data Engineer**: manages training data pipelines, [[Synthetic Data]] generation, data quality, and [[AI Training Data Collection]] governance - **AI Platform Engineer**: operates infrastructure for [[AI Inference]], model serving, [[Model Routing]], and [[AI Cost Management]] ### Strategic - **AI Product Manager**: defines AI-powered features; balances capability with [[AI Risks and Fears|risk]]; manages the [[AI and Trust|trust]] tradeoff - **AI Ethics/Safety Lead**: owns [[AI Ethics]], [[AI Safety]], [[AI Alignment]], and [[Responsible AI]] compliance - **AI Governance Officer**: enforces [[AI Usage Policy]], [[AI Data Security]], and regulatory compliance (e.g., [[EU AI Act]]) ### Operational - **AI Trainer/Evaluator**: handles [[Reinforcement Learning From Human Feedback (RLHF)|RLHF]], evaluation datasets, and quality benchmarking - **AI Change Manager**: drives [[Team AI Onboarding]], [[AI Literacy]] programs, and adoption metrics ## How Roles Map to the AI Stack | Layer | Primary Owner | |-------|--------------| | Model selection and fine-tuning | AI/ML Engineer | | Context and prompt systems | Prompt/Context Engineer | | Agent harness and skills | AI Platform + Prompt Engineer | | Data pipelines | Data Engineer | | Governance and policy | AI Governance Officer | | Safety and ethics review | AI Ethics Lead | | Product decisions | AI Product Manager | ## Key Principles - **No single person owns "AI"**: it spans engineering, product, legal, and operations - Smaller teams collapse roles; a startup might have one person covering all technical roles - The [[AI and the Shifting Role of Developers|developer role is shifting]] from code crafting to architecture and review - [[Shadow AI]] emerges when governance roles are absent or understaffed ## References - ## Related - [[AI Governance]] - [[AI Implementation Roadmap]] - [[Team AI Onboarding]] - [[AI and the Shifting Role of Developers]] - [[Enterprise AI Deployment]] - [[AI Engineering]] - [[AI for Enterprise Leaders]]