AWS Professional Services redesigned its consulting delivery from the ground up using AI agents, compressing timelines from months to days by building AI into the workflow rather than layering it on top.

What happened

AWS Professional Services created APEX (Agentic AI ProServe Experiences), a team that built the ProServe Delivery Agent—a multi-agent system that handles requirements, architecture validation, implementation, security review, testing, and deployment alongside human consultants. The Delivery Agent now works on customer engagements globally and represents how ProServe delivers at scale, not as a pilot program.

Why it matters

Traditional consulting workflows involve long handoffs and delays between discovery, architecture, implementation, and testing phases. By treating AI agents as the foundation rather than an assistant, and restructuring work around what agents do well (generating code, running tests, self-correcting output), AWS freed consultants from non-coding overhead like documentation and status reporting so they can focus on high-judgment decisions. This shift allows customer projects to move at the pace that AI-native development demands.

What to watch

On customer engagements, the governing principle is "humans provide intent, AI creates, humans verify." One customer used the Delivery Agent to adopt Amazon Application Recovery Controller's Region Switch functionality, with Kiro and the Delivery Agent compressing weeks of backlog creation into hours and accelerating code delivery by 60%, while enforcing consistent quality.