groundcover Adds Azure Support for AI Observability Tooling

groundcover announced that its Agent Mode observability offering now supports Microsoft Azure, positioning the update as a way for teams to monitor production AI and LLM workloads on Azure.

Azure support for Agent Mode

In a company blog post, groundcover said it has brought “Observability for Production AI” to Microsoft Azure through Agent Mode. The post says AI and LLM observability data can be queried through an included MCP server, and describes the deployment as a bring-your-own-cloud model.

A Business Wire release from groundcover similarly says the company updated Agent Mode with native Microsoft Azure support and planned to demonstrate the product at Microsoft Build 2026 at Fort Mason Center, booth G-222. The release describes the product as “now live on Microsoft Azure,” indicating availability rather than a future roadmap item.

The announcement matters for organizations that are trying to apply familiar production monitoring practices to AI systems. As companies deploy LLM-backed applications, observability often needs to cover not only infrastructure health, but also application behavior, model interactions, latency, and operational context. groundcover’s framing focuses on making that information available for Azure-based environments.

MCP server access for observability data

According to groundcover’s blog post, the Azure release includes an MCP server that allows AI and LLM observability data to be queried. MCP, or Model Context Protocol, is commonly used to connect AI systems with external tools and data sources, although groundcover’s sources specifically describe its role here as a way to query observability information.

The company’s announcement does not, in the provided excerpts, specify the full set of telemetry fields exposed through the MCP server or the supported query patterns. It does state that the included server is part of the Agent Mode release for Azure.

Bring-your-own-cloud deployment

Both groundcover’s blog excerpt and the Microsoft Marketplace listing emphasize customer-controlled infrastructure. The Marketplace page lists groundcover as a full-stack observability platform for Azure workloads that collects logs, metrics, and traces using eBPF while keeping data in the customer cloud.

That deployment detail is important for regulated or security-conscious teams evaluating observability vendors. Keeping telemetry within a customer cloud environment can reduce some data movement concerns, though the sources provided do not make specific compliance claims beyond describing where data is kept.

The Microsoft Marketplace listing also describes groundcover as supporting Azure workloads and collecting core observability data types: logs, metrics, and traces. Those are standard inputs for infrastructure and application monitoring, and the Marketplace page ties them to groundcover’s eBPF-based collection approach.

Context at Microsoft Build

groundcover tied the release to Microsoft Build 2026. The Business Wire release says the company would showcase the enhanced Agent Mode at the event and identifies a booth location at Fort Mason Center.

The announcement fits a broader pattern of observability vendors adapting their products for AI application operations, but the substantiated details here are narrower: groundcover says Agent Mode now supports Azure, includes an MCP server for querying AI and LLM observability data, and uses a bring-your-own-cloud model. Microsoft Marketplace separately lists the product as an Azure observability platform collecting logs, metrics, and traces while keeping data in the customer cloud.

For Azure users, the practical takeaway is that groundcover is making its AI observability tooling available in Microsoft’s cloud ecosystem, with distribution and positioning supported by its Microsoft Marketplace presence.