9. AI Architect

Evidence label: Role family, usually senior or staff level. AI Architect, Applied AI Architect, and Enterprise AI Architect postings share system-level decision ownership but differ in how hands-on they are.

What they do

Design enterprise AI systems at the system level. Where AI engineers ship features, AI architects define the patterns, platforms, and decision frameworks that multiple teams use. They make the technology choices that others live with for years.

Responsibilities

• Define AI technology strategy and stack decisions (which models, which frameworks, which providers)
• Design enterprise AI reference architectures
• Set standards for AI system observability, security, and governance
• Evaluate build vs. buy decisions for AI capabilities
• Ensure AI systems are scalable, cost-effective, and auditable
• Bridge between business requirements and technical AI implementation

Required skills

• Deep experience across AI/ML stack (models, infrastructure, MLOps)
• Strong communication skills (presenting to C-suite, working with legal/compliance)
• Understanding of cloud provider AI offerings (AWS Bedrock, Azure OpenAI, Vertex AI)
• Security and compliance awareness (GDPR, AI Act, SOC2)
• Experience designing distributed systems at scale

Entry paths

Senior or staff AI engineer moving into cross-team architecture. Another path is a cloud or platform architect who adds model, evaluation, and AI-governance depth.