Build 2026: From observability to ROI for AI agents on any framework
Microsoft Foundry's agent tracing and evaluations hit GA and now extend to any agent framework and deployment target.
“Shipping an AI agent is the easy part. Keeping it accurate, safe, and accountable in production is where teams get stuck.”
At Build 2026, Microsoft announced that tracing and evaluations in Foundry reached general availability and are expanding to support any agent framework, any deployment target, and the full Agent DevOps loop through to ROI dashboards. This matters because production observability and evaluation are emerging as the critical bottleneck for trustworthy, non-deterministic AI agents, and Microsoft is positioning Foundry as a framework-agnostic standard for it.