Azure Updates: Agent Factory; AKS Automatic; Container Storage

Microsoft CVP Yina Arenas looked at capabilities around Agent Factory. Customers are increasingly using Azure AI Foundry to connect and coordinate different agents, for instance aligning customer service agents and retrieval agents. 

In the past, Microsoft used OData to standardize data API access and helped developers leverage OpenTelemetry to get common ground for observability. “Today, customers in Azure AI Foundry are looking for flexibility without vendor lock-in. The same pattern is now unfolding with AI agents. Proprietary, closed ecosystems create risk if agents, tools, or data can’t interoperate, causing innovation to stall and an increase in switching costs,” Arenas wrote.

To create a “lingua franca” for agents to share tools and context, Microsoft supports common protocols such as Agent2Agent (A2A) and Model Context Protocol (MCP). It also supports interop across frameworks with AutoGen, LangGraph, Semantic Kernel, and CrewAI. 

CVP Brendan Burns announced the general availability of AKS Automatic. Microsoft is working to enable configured and optimized AKS capabilities with intelligent defaults. AKS Automatic is intended to reduce the learning curve for new Kubernetes users, free up resources, mitigate security, and cut down on reliability problems associated with misconfigurations. 

According to Burns, AKS Automatic enables dynamic scaling for both pods and nodes with Horizontal Pod Autoscaler (HPA), Vertical Pod Autoscaler (VPA), and KEDA for event-driven scaling. Additionally, it introduces automated node provisioning with Karpenter, an open-source Kubernetes autoscaler that dynamically scales clusters based on demand.

Microsoft VP Aung Oo looked at how to accelerate AI and databases with Azure Container Storage. A growing number of organizations are running stateful workloads, including messaging queues and AI inferencing. Microsoft deployed a new release of Container Storage—v2.0.0—that offers seven times greater IOPS and four times less latency.