CopilotKit raises $27 million to help developers deploy native AI agents for apps

Many companies today provide AI simply as a chatbot within their apps: you write (or dictate) what you want it to do, and the AI ​​bot goes and tries to do it. However, the experience tends to feel heavy. A text-based UI doesn’t always translate into a smooth experience, for example, if you want to use a travel app to book an entire itinerary but have to scan through blocks of text.

According to the founders CopilotKit,This approach does not make the most of what AI agents and LLMs can do. The company’s co-founders, Atay Barkai (pictured above, right) and Uli Barkai (pictured above, left), believe the way forward is to enable agents to live inside applications, understand what users are doing, take actions, and show useful interfaces rather than just returning long blocks of text.

The company’s popular AG-UI protocol is aimed at the first part of this solution. The widely adopted open source protocol standardizes how AI agents connect and communicate with user interfaces (such as a web browser or app), providing features such as streaming chat, front-end tool calls, and state sharing to enable human-in-the-loop functionality. Essentially, AG-UI provides developers with the framework and tools needed to deploy AI agents within their applications.

CopilotKit is also building an enterprise toolset on top of AG-UI, adding support, self-hosted publishing features, and other essential offerings for companies considering building agents into their products. To bring this toolkit to market, the Seattle-based startup has raised $27 million in a Series A round led by Glilot Capital, NFX, and SignalFire, TechCrunch has learned exclusively.

The flexible user interface is a particular selling point. CEO Atay Barkai told TechCrunch that developers can use the startup’s framework to provide specifications and key elements for dynamic user interfaces, which an AI agent can then use to create contextually appropriate user interfaces.

“The agent can respond to you, not just through blocks of text, but through interactive user interfaces defined by your own company,” Atay explained. “For example, if a user asks for a breakdown of revenue by category, instead of having this kind of big, impenetrable paragraph, you get a pie chart, which is your own design of the pie chart that the user can interact with (…) so all your agents can, very trivially, talk to the UI and use the component catalog and show that to users.”

Atai also noted that the CopilotKit toolkit gives developers full control over how much their AI agent can change the UI, to the point where they can choose to have the interface be “pixel perfect” or just provide broad building blocks that the AI ​​can piece together as required.

The funding comes after a period of strong adoption of both AG-UI and CopilotKit. A protocol that works alongside the widely adopted protocol Model Context Protocol (MCP) and Agent2Agent (A2A) protocol.Today, it is supported by leading AI infrastructure providers such as Google, Microsoft, Amazonand oraclein addition to popular frameworks such as langshen, Mastered, PydanticAIand I know.

Atay said that CopilotKit and AG-UI (the company’s strongest claim to ecosystem importance) are seeing millions of installations weekly, and that a significant portion of Fortune 500 companies are using the protocol and the startup’s tools in production. Meanwhile, CopilotKit counts enterprise heavyweights such as Deutsche Telekom, Docusign, Cisco and S&P Global as institutional clients.

To capitalize on this growing interest, the company is also launching CopilotKit Enterprise Intelligence, a self-hostable offering that brings together a number of infrastructure features to fully deploy agents within applications.

CopilotKit faces stiff competition in the enterprise agent tools market. The Vercel cloud platform is open source Artificial Intelligence SDK It helps developers build AI applications with similar capabilities, and User Interface Assistant Provides components for building AI chat interfaces. Meanwhile, OpenAI’s Apps SDK is also an option for building richer interfaces, but only within ChatGPT.

CopilotKit differs from those offerings because it takes a horizontal, enterprise-friendly approach rather than a vertically integrated approach, Atay says. Rather than providing an integrated AI platform, CopilotKit aims to support any agent framework, cloud provider, or backend an organization is already using.

“If there are two things we hear in almost every enterprise conversation, it’s that companies want optionality and they want self-hosting,” he said. “Maybe they’re already using Google Groups, Amazon, Oracle, Microsoft, LangChain, Mastra. They want optionality, they want self-hosting, and those are two things they don’t really get in the Vercel stack.”

It will be important to maintain this open position. Companies that rely on their own open source infrastructure often face a tension, which is that they want their technology to remain a neutral standard, but they also need to build businesses on top of it. But Atay said AG-UI is a fully open protocol, and that CopilotKit’s commercial product is intended to strengthen the enterprise open source stack, not replace it.

“They are very complementary,” added Ollie, the startup’s head of growth. “Our strategy is to be the default in the ecosystem, and then monetize the bigger companies.” “So it’s in our interest that open source is the best it can be, and 95% of users can start building and get going without paying anyone or talking to anyone.”

The company currently has about 25 employees and plans to use the new funding to grow its team.