Schneider Electric, Vodafone, and monday.com Share Agent Operations Infrastructure Implementation Cases in Europe and the Middle East

AI agent adoption in Europe and the Middle East is focusing on building central platforms and operations infrastructure rather than consumer chatbots. Many companies have dozens of prototypes but lack the infrastructure to operate them consistently, struggling to transition to production environments.

Key Company Cases

• Schneider Electric: Operates an internal AI hub of 350 people, supporting more than 60 agents through an LLMOps framework centered on Observability, evaluation, and deployment.
• monday.com: Redesigned its AI assistant Sidekick from a single general-purpose agent into a hierarchical system including sub-agents, restricted tools, and sandboxes. This change follows production lessons that adding tools causes performance degradation.
• Vodafone: Built two production assistants, the LangGraph-based Insight Engine and Enigma, and performs monitoring and improvement tasks via LangSmith.

Regional Patterns and Implications

35% of surveyed organizations cited enterprise-wide agent platforms or Control Planes, on which business unit agents run, as their primary use case. This aims to prevent individual teams from repeatedly building the same foundation and allows central AI hubs to support the entire lifecycle.

Additionally, 18% of organizations are securing ROI in regulated back-office tasks with clear per-transaction costs, such as insurance claims, underwriting, and invoicing. In a reality where agent prototyping is easy but operations are difficult, the absence of this infrastructure layer is identified as a major bottleneck.