# 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]]