From Change Data to Live Context: Serving CDC Streams to AI Agents over MCP

Every enterprise wants AI agents that act on what's happening right now, but most agents reason over stale snapshots — nightly exports, cached lookups, or vector stores rebuilt hours after the underlying data changed. In a regulated healthcare environment, that lag isn't just a quality problem: acting on out-of-date member, claims, or eligibility data carries real compliance and member-impact risk.

This session walks through how we turned our existing change-data-capture (CDC) pipelines into a real-time context layer that AI agents query directly over the Model Context Protocol (MCP) — without adopting a packaged context engine. Running self-managed Confluent Platform across GCP and Azure, we use Debezium to capture row-level changes from operational databases, enforce structure with Schema Registry and Avro under backward-compatibility rules, and materialize that stream into a queryable serving layer that always reflects current state. A custom MCP server exposes this layer as governed tools and resources, so agents retrieve fresh, contextual data on demand instead of relying on brittle polling or stale embeddings.

We'll go deep on the parts that actually matter in production: topic and schema design that keeps the context layer evolvable, how we handle PII/PHI through field-level governance and access scoping at the MCP boundary, and how we trace and audit every agent query end to end using OpenTelemetry context propagated through Kafka headers.

Attendees will leave with a concrete architectural blueprint for serving streaming data to agents, a clear view of why CDC is a natural foundation for real-time context, and a practical checklist for doing it safely in a regulated domain: schema governance, access control, observability, and the failure modes to plan for. Whether you're building agentic systems or just trying to keep your AI grounded in current reality, you'll walk away knowing how to make your data the freshest thing your agents touch.