Build Interactive Agents with Generative UI

The Batch / 5/7/2026

💬 OpinionDeveloper Stack & InfrastructureTools & Practical UsageModels & Research

Read original →

共有:

Key Points

The article promotes a short, intermediate-level course focused on building interactive agents using a “Generative UI” approach.
It is led by instructor Atai Barkai and positioned as a practical, hands-on training session lasting about 1 hour and 25 minutes.
The course title emphasizes combining agent behavior with generative, UI-driven interactions rather than treating the UI as a static layer.
While the page content provided is mostly promotional/landing-page material, the framing indicates a focus on developers creating agent experiences end-to-end.

Short CourseIntermediate1 Hour 25 Minutes

Build Interactive Agents with Generative UI

Instructor: Atai Barkai

Enroll Now

Intermediate
1 Hour 25 Minutes
8 Video Lessons
5 Code Examples
Instructor: Atai Barkai
CopilotKit

What you'll learn

Understand the three approaches to building agent interfaces on the Generative UI Spectrum: Controlled, Declarative, and Open-Ended Gen UI, and when to use each.
Build a fullstack agent app by connecting a LangChain agent to a React frontend using CopilotKit and the AG-UI protocol, with agents that render charts, cards, and forms on demand.
Use MCP Apps to connect your agent to third-party applications, then go beyond the chat window and build products where the agent and user work on the same data, side by side.

About this course

Introducing Build Interactive Agents with Generative UI, taught by Atai Barkai, co-founder of CopilotKit, the team behind the AG-UI protocol.

Most agents today still talk to users in plain text. But users don’t just want to read a response, they want something to see and act on. This course teaches you to build that interactivity: a fullstack agent interface where the agent renders charts, forms, whiteboards, and interactive components on demand.

You’ll work across the full Generative UI Spectrum, from hand-crafted components you control precisely, to declarative layouts the agent assembles from building blocks, all the way to open-ended experiences powered by MCP Apps. You’ll finish by building a canvas application where the agent works alongside the user on shared data. By the end, you’ll have a production-ready fullstack agent built on CopilotKit and AG-UI, an open protocol with first-party integrations across LangGraph, Google, AWS, Microsoft, and more.

In detail, you’ll:

Learn the three pillars of the Generative UI Spectrum and when to use each.
Connect a LangChain Deep Agent and a Google ADK Agent to a React frontend using CopilotKit and AG-UI.
Build Controlled Generative UI: define custom components like pie charts and flight cards that the agent renders on demand.
Use A2UI, an open spec co-developed with Google, to let your agent assemble layouts from a catalog of reusable building blocks.
Connect your agent to MCP Apps and explore fully open-ended interfaces the agent creates from scratch.
Build a canvas app where the agent and frontend share live state, beyond the chat window.

Throughout the course, you’ll build on the open-source AG-UI protocol — so these skills travel with you across the agentic ecosystem.

Who should join?

Developers building AI agents who want to create richer, more interactive user experiences. Comfort with React and Python is recommended, along with basic familiarity with AI agents or LLM application development. Useful for fullstack developers, AI engineers, and frontend-leaning builders shipping agent features into production apps.