Elastic announced a new integration with Azure Artificial Intelligence Foundry that adds observability for agentic Artificial Intelligence applications and large language models. The integration surfaces real-time insights for site reliability engineers and developers into model usage, generative Artificial Intelligence workloads and agentic Artificial Intelligence behavior. Elastic positions the integration as a way to help teams build, monitor, and optimize intelligent agents on Azure Artificial Intelligence Foundry with improved reliability, efficiency and guardrails.
The release addresses common operational challenges organizations face as they deploy agentic Artificial Intelligence in production, including uncontrolled token usage, latency bottlenecks and compliance blind spots. Elastic provides pre-built dashboards that present a unified view of model usage, performance, costs and content filtering so teams can identify bottlenecks, optimize configurations and understand cost drivers in real time. The company says these capabilities let organizations scale Artificial Intelligence applications faster without sacrificing reliability, compliance or budget control.
Executives quoted in the announcement emphasized operational clarity and safeguards. “Agentic Artificial Intelligence is only as strong as the models and infrastructure that power it,” said Santosh Krishnan, general manager, observability & security at Elastic, noting that the integration helps teams fix performance bottlenecks and understand cost drivers. Amanda Silver, corporate vice president at Microsoft Azure CoreArtificial Intelligence, said the integration delivers real-time visibility into token usage, latency and costs, and adds built-in safeguards for models hosted in Azure Artificial Intelligence Foundry. The Elastic Azure Artificial Intelligence Foundry integration is available in tech preview on Elastic Observability.
