​

Hebo AI

​

Introduction

Building production-ready conversational agents requires fast iteration across engineering, product, and business teams. In practice, iteration is slow, even small changes often require code updates and lengthy re-deployments. Hebo addresses this by providing a shared platform for building, operating, and improving AI agents:

• A gateway with standardized OpenAI-compatible API across LLMs
• Composable primitives for RAG, tools, and memory
• Built-in observability & evaluations focused on real conversation flows

Wrapped into a simple user interface for operational teams to define agent behavior, observe conversations and optimize knowledge. Every component is opt-in and designed to scale with agent complexity, from simple prompts to production systems.

​

Component Status

• Gateway: v0.1 (live)
• MCP AIKit: v0.1 (live)
• Evals: v0.1 (live)
• RAG / Tools / Memory stack: WIP
• Observability: WIP
• Knowledge Editor: WIP

​

Getting Started

1. Register your Account: Visit console.hebo.ai to create an account.
2. Create an Agent: The on-boarding flow will guide you through creating an agent and selecting a default model.
3. Execute API Calls: Use your API keys to interact with your agents programmatically.

​

Support

Join us on Discord for support and discussions: Discord Channel

​

Contributing

Hebo is built in the open with an Open-Source license that keeps Hebo free to use for everyone while discouraging harmful free-riding. We’re still at the very beginning and welcome contributions from the community. To get started:

1. Open an issue from our issue tracker.
2. Open a pull request with your proposed changes.
3. Ping us on our discord channel.

We look forward to your contributions!