The Cisco Outshift Platform Engineering team developed JARVIS, an AI Platform Engineer designed as a distributed Multi-Agent System (MAS) to unlock a 10x productivity boost. JARVIS is orchestrated using LangGraph for scalable and deterministic agent workflows and connected through the AGNTCY Agent Connect Protocol (ACP), enabling seamless agent-to-agent collaboration across systems. It integrates with platform knowledge bases, automates many developer tasks, simplifies Kubernetes deployments, surfaces agentic AI capabilities in familiar developer interfaces, and drives adoption while enhancing platform usability and responsiveness. The development of JARVIS was grounded in AGNTCY's Four-Phase approach to building resilient multi-agent systems on the Internet of Agents, which includes Discover, Compose, Deploy, and Evaluate phases. JARVIS has delivered significant productivity gains for Platform Engineering at Outshift, including tasks that previously took a week now completed in under an hour, provisioning resources instantaneous, and eliminating back-and-forth communication between developers and Platform Engineering.