In this episode of IoT Innovators, we focus on what it takes to move AI from controlled demos into real operational environments. Brandon Kooman sits down with Rik de Ruiter and Andrey Matveenko to break down how SAS Event Stream Processing (ESP) and Model Context Protocol (MCP) work together to make that shift practical.
The discussion starts with a clear challenge. Industrial systems are complex, fast moving, and often noisy. Alerts pile up, equipment degrades, and experienced operators leave with critical knowledge. Natural language helps teams interact with these systems, but without connection to live data, AI produces answers that look right but aren’t grounded in reality.
Two core capabilities drive the solution:
- SAS Event Stream Processing: ESP provides the execution layer. It processes real time data at scale and adds the control needed in production environments. Teams get consistency, traceability, and clear visibility into how decisions are made, which is critical in regulated and safety focused settings.
- Model Context Protocol: MCP changes how AI agents interact with operational systems. Instead of relying on rigid API calls, agents identify the right tools, apply them in context, and explain results clearly. It builds on existing systems, so teams extend what they already run instead of rebuilding integrations
Beyond the architecture, the episode stays focused on how to start. The guidance is simple. Pick one use case, connect a small set of tools, prove value quickly, then expand. This approach avoids overbuilding and keeps progress measurable.
You leave with a clear view of how MCP and ESP work together, where AI delivers value today, and what steps matter most when moving from prototype to production.
Link to the Video
