CopilotKit Raises $27 Million to Expand App-Native AI Agent Tools

Many software companies still integrate artificial intelligence as chat-based assistants. However, this approach often creates a fragmented user experience. For instance, users may need to read long text responses to complete simple actions, which slows down workflows.

CopilotKit, a Seattle-based startup, is working on a different approach. Instead of relying only on chat interfaces, the company focuses on embedding AI agents directly into applications. These agents can understand user context, take actions, and present results through interactive interfaces.

Building AI Agents Within Applications

CopilotKit was founded by Atai Barkai and Uli Barkai. The founders believe AI agents should operate within the structure of an application rather than outside it. As a result, users can interact with features visually instead of relying on text-heavy responses.

To support this, the company developed AG-UI, an open-source protocol. This protocol defines how AI agents connect with user interfaces such as web apps and browsers. It also enables features like real-time interaction, tool execution, and shared state between users and agents.

In simple terms, AG-UI provides developers with a structured way to integrate AI agents into their products.

Enterprise Toolkit and Product Expansion

On top of AG-UI, CopilotKit is building an enterprise-focused toolkit. This includes support for self-hosted deployments and additional infrastructure required by large organizations.

To expand this offering, the company raised $27 million in a Series A funding round. The round was led by Glilot Capital, NFX, and SignalFire.

The funding will help CopilotKit scale its enterprise product and grow its team, which currently has around 25 employees.

Focus on Interactive User Interfaces

One of the company’s key features is its flexible interface system. Developers can define UI components that AI agents can use dynamically.

For example, instead of returning a long text explanation, an AI agent can display structured outputs such as charts or dashboards. This makes the interaction clearer and easier to use.

In addition, developers can control how much freedom the AI has in modifying the interface. They can either enforce strict design rules or allow more flexibility based on use cases.

Adoption Across AI Ecosystem

CopilotKit reports strong adoption of its AG-UI protocol. The framework is compatible with widely used technologies such as Model Context Protocol (MCP) and Agent2Agent (A2A).

It is also supported by major cloud providers, including Google, Microsoft, Amazon, and Oracle. Furthermore, popular AI development frameworks like LangChain, Mastra, and PydanticAI integrate with the protocol.

According to the company, AG-UI records millions of installs each week. Several large enterprises, including Deutsche Telekom, DocuSign, Cisco, and S&P Global, are already using its tools in production.

Competition and Market Positioning

The market for enterprise AI agent tools is becoming increasingly competitive. Companies like Vercel offer open-source AI SDKs, while assistant-ui focuses on building chat-based interfaces. OpenAI’s Apps SDK also enables richer experiences, though within its own ecosystem.

CopilotKit positions itself differently. Instead of offering a full-stack platform, it supports multiple frameworks and cloud providers. This approach allows enterprises to integrate AI agents without changing their existing infrastructure.

The company emphasizes two key requirements from enterprise clients: flexibility and self-hosting. These features are central to its product strategy.

Balancing Open Source and Commercial Growth

CopilotKit operates both an open-source protocol and a commercial enterprise product. This creates a common challenge—maintaining openness while building a business model.

The company states that AG-UI will remain fully open. Meanwhile, its enterprise offerings are designed to enhance the open-source stack rather than replace it.

Its strategy focuses on broad adoption of the open protocol, while monetizing enterprise-grade features for larger customers.