Build a chat agent on Chromia

You get boilerplate code for a minimal AI-powered chat agent running on Chromia. Treat it as a starting point for experiments: try out different agent strategies, change how memory is managed, and plug in other AI models.

What is Chromia?

Chromia is a Relational Blockchain, meaning it merges relational database features with a blockchain to make building decentralized applications (dapps) simpler. Whether you come from enterprise software or game development, you can build dapps using patterns you already know.

One of Chromia's key components is Rell, a concise language for both blockchain and database work. With Rell you write efficient, secure, expressive applications in little code, and you keep the flexibility of a relational database.

For a broader look at the platform, read the Chromia overview.

About the dapp

Your starting point is a boilerplate chat agent driven by an AI model. It keeps context in short-term and long-term memory and retrieves it later, so its replies stay dynamic and aware of the conversation so far.

From there, you can try different AI strategies, adjust how memory is handled, and swap between AI models (such as Groq and OpenAI) to change how the agent behaves.

The demo below shows the chat agent running:

What will I learn?

When you finish, you will have hands-on experience with:

• Setting up and running a dapp built on Chromia.
• Trying out memory management strategies that improve contextual accuracy.
• Connecting several AI models and tailoring how the agent responds.
• Writing backend logic in Rell, including queries and transactions.
• Reshaping the chat agent for your own requirements or projects.

You end up with a working boilerplate that you can extend and change to build your own interactive chatbot dapp.

Repository link

The full source code is available in the chat agent course repository.