Building the internet of agents

Recently, Nasiko hosted a community workshop as part of Project NANDA: The Internet of AI Agents, bringing together developers, researchers, and builders to explore the emerging infrastructure layer powering the next generation of AI systems.

The session focused on a shift that is rapidly taking shape across the AI ecosystem.

For the past few years, most AI applications have existed as isolated systems—agents interacting with users, tools, and data within a single environment. The next phase is different. Agents are increasingly expected to communicate, collaborate, coordinate, and transact across platforms, organizations, and execution environments.

Enabling this future requires more than better models.

It requires infrastructure.

During the workshop, Justin Elof Johnson, Developer Experience Engineer at Nasiko, shared perspectives on the evolution of the AI ecosystem and demonstrated how agent networks can move beyond experimentation toward production-ready deployments.

Participants explored several foundational concepts behind the emerging Internet of Agents:

• Agent Control Planes that coordinate, manage, and govern large populations of agents.

• Interoperability frameworks that allow agents built on different systems to communicate and work together.

• Production infrastructure that supports reliability, discovery, observability, and lifecycle management at scale.

• Open ecosystems where developers can build, test, and deploy agent-native applications.

The workshop also featured a hands-on demonstration of NEST, an open sandbox environment where agents can discover services, interact with one another, and participate in shared agent ecosystems.

What made the session particularly encouraging was the level of engagement from the developer community. Discussions quickly moved beyond model capabilities and into deeper questions around agent identity, trust, governance, discovery, coordination, and network design.

These are the challenges that will define whether agent systems remain isolated experiments or evolve into a truly interconnected ecosystem.

At Nasiko, we believe the future of AI will be increasingly networked. As agents become participants in larger systems, enterprises and developers will need the infrastructure that enables those agents to operate reliably, securely, and at scale.

The NANDA workshop was another step toward building that future—one where agents are not only intelligent, but discoverable, interoperable, and capable of working together across the broader Internet of Agents.

We're excited to continue supporting the developer community and advancing the infrastructure needed for the next generation of agent-native applications.