At our recent Nashua Cloud .NET User Group meetup, I delivered a session titled “Connecting Multi-Cloud AI Agents with Semantic Kernel,” exploring how organizations can orchestrate intelligent agents across Microsoft, AWS, and Google ecosystems. As AI adoption accelerates, enterprises increasingly need interoperability, portability, and flexible agent orchestration—and this session focused on exactly how to achieve that.
We began by breaking down agent capabilities within the Microsoft ecosystem: Copilot Studio for no-code conversational agents, Azure AI Foundry for enterprise-grade evaluation and orchestration, and frameworks like AutoGen, Semantic Kernel, and the emerging Agent-Framework for building customizable, tool-driven, multi-agent systems. Each offers unique strengths—ranging from memory, planning, and tool chaining to human-in-the-loop workflows and cross-app integration. Muti-Cloud Agents
From there, we explored the agent platforms in AWS and Google Cloud, including AWS Agents for Bedrock and Google Agent Builder. Both platforms emphasize grounding, tool execution, and enterprise connectors, making multi-cloud agent orchestration more practical than ever for real-world systems. Muti-Cloud Agents
A live demo showed how Semantic Kernel can act as the “glue” across clouds—coordinating planning, tools, skills, and memory while delegating execution to Microsoft, AWS, or Google agent endpoints. The audience also saw how SK simplifies building portable agents that can operate in hybrid or multi-cloud environments without rewriting the core intelligence layer.
Thank you to everyone who joined, asked thoughtful questions, and contributed to a great discussion.
The Recording and sample codes are available here:
Watch the session:https://youtu.be/Tg9vjsLyLdQ
Sample Code: https://github.com/nhcloud/agentcon-workshop
If you missed the session, join us at the next Nashua Cloud .NET User Group (NashuaUG) meetup to continue exploring practical, real-world AI engineering.