Dapr Agents v1.0: CNCF Ships Production-Ready AI Agent Framework at KubeCon
The Cloud Native Computing Foundation has announced the general availability of Dapr Agents v1.0, a Python framework for building resilient, production-ready AI agents. The announcement was made at KubeCon + CloudNativeCon Europe 2026 in Amsterdam on March 23.
Dapr Agents provides the infrastructure to move AI agents from prototypes to production: durable long-running workflows with automatic retries and failure recovery, persistent state management across 30+ databases, secure communication using SPIFFE identity, multi-agent coordination and messaging, and built-in observability. The framework also offers the flexibility to switch language model providers without code changes.
The 1.0 release is the result of a yearlong collaboration between NVIDIA, the Dapr open-source community, and end users. ZEISS Vision Care presented a real-world implementation at KubeCon, using Dapr Agents to extract optical parameters from unstructured documents.
KubeCon Europe 2026 also featured a brand-new Agentics Day: MCP + Agents co-located event — a half-day deep dive into the MCP ecosystem and agentic workflows — signaling how central agent infrastructure has become to the cloud-native ecosystem.
GitHub: https://github.com/dapr/dapr-agents
Announcement: https://www.cncf.io/announcements/2026/03/23/general-availability-of-dapr-agents-delivers-production-reliability-for-enterprise-ai/
← 返回所有文章
Dapr Agents provides the infrastructure to move AI agents from prototypes to production: durable long-running workflows with automatic retries and failure recovery, persistent state management across 30+ databases, secure communication using SPIFFE identity, multi-agent coordination and messaging, and built-in observability. The framework also offers the flexibility to switch language model providers without code changes.
The 1.0 release is the result of a yearlong collaboration between NVIDIA, the Dapr open-source community, and end users. ZEISS Vision Care presented a real-world implementation at KubeCon, using Dapr Agents to extract optical parameters from unstructured documents.
KubeCon Europe 2026 also featured a brand-new Agentics Day: MCP + Agents co-located event — a half-day deep dive into the MCP ecosystem and agentic workflows — signaling how central agent infrastructure has become to the cloud-native ecosystem.
GitHub: https://github.com/dapr/dapr-agents
Announcement: https://www.cncf.io/announcements/2026/03/23/general-availability-of-dapr-agents-delivers-production-reliability-for-enterprise-ai/
评论