Bringing a DeepSeek R1 LangGraph Agent Into The Real World Using CopilotKit

Hey everyone!

I'm a DevRel at CopilotKit, an open-source framework for agentic UI and in-app agents.

I wanted to share a video we produced to show how easy it is to give your app AI superpowers in just minutes.

The DeepSeek R1 LangGraph agent demo is powered by one simple hook, the useCoAgent hook (https://docs.copilotkit.ai/reference/....
Then, viola, you are cooking with gas so to speak!

Resources:

• CopilotKit GitHub Repo

• Ollama Deep Researcher source code

Follow CopilotKit on Twitter and say hi, and if you'd like to build something cool, join our Discord.

Comments

[PRANKUR PANDEY CopilotKit]

This looks interesting

[Nathan Tarbert CopilotKit]

Thanks, Prankur, I'd love to get your feedback.

[K Om Senapati]

Amazing

[Nathan Tarbert CopilotKit]

Thanks, K OM!

[Md Abid Hussain]

Great 🔥🔥

[Nathan Tarbert CopilotKit]

Thanks, MD!

[Cezar Lamann]

It looks good, but could you please post the link to your github repo for this project?

[Nathan Tarbert CopilotKit]

Hey Cezar, thanks for your feedback!

I've added the links to the description

[David]

This looks cool!

[Nathan Tarbert CopilotKit]

Thanks, yes, it is very cool + really simple to drop into your app.

[Abdul Rehman Baber]

Beside copilot hook everything seems not so special.

The main problem for local development of a deep research type module is not any of this but the how scrap, read bunch of content available on internet with efficiency and give it to llm effectively and maintain the context and state for the "important, relevant information" through out. Use tavily without orchestration and see its outputs, its basic

This is just a stunt nothing more. Sorry

[John Byrne]

This looks cool! Great article 🔥

[R Jones]

Nice post