About This Architecture

Unified KPI metric collection platform on Azure ingests telemetry from multiple AI agents (IES MINT, Placemat, MSD) via a Telemetry SDK into Azure Event Hub, where schema validation and dead-letter queuing ensure data quality. Raw events flow to Azure Data Explorer (Kusto) and are transformed by Azure Functions ETL into curated KPI tables via materialized views, which feed Power BI dashboards, Placemat analytics, and alerting systems. Governance layers enforce data contracts, metric definitions, and RBAC, enabling agent teams and platform teams to collaborate on reliable, auditable KPI measurement across the organization. Fork this diagram to customize agent sources, add custom ETL logic, or integrate additional analytics tools while maintaining schema governance and access control.

People also ask

How do I build a unified KPI metric collection platform on Azure that ingests from multiple agents, validates schemas, and serves analytics dashboards with governance?

This diagram shows a four-layer Azure architecture: Layer 1 ingests telemetry via Event Hub with schema validation and dead-letter queuing; Layer 2 stores raw events in Data Explorer and transforms them into curated KPI tables using Azure Functions; Layer 3 serves Power BI dashboards and alerts; Layer 4 enforces data contracts and RBAC. This pattern ensures data quality, scalability, and complianc