Voiceflow is the fastest way to build a high complexity agent with your team, and our documentation is organized around your journey using Voiceflow.

Check out our Quick Start Guide to understand the main components of Voiceflow, and here are the main sections of our documentation.

Building Agents

This is all about creating your MVP agent in Voiceflow. Our recommendation is to start with a simple use case, deploy it to production, then start improving and expanding it as you learn how your customers interact with it.

1. Platform fundamentals: Deep dives into the core parts of Voiceflow.
2. Understand: How your agent understands what the user is saying and how to fix issues.
3. Decide: How your agent determines where to route the user after it understands what they are asking.
4. Response: How your agent uses AI, the Knowledge Base, and other factors to craft a response.

Deploying Agents

This is about actually launching your Agent to production. Either using the Dialog API to connect to a custom interface to build a native co-pilot for your app or an agent in WhatsApp, MS Teams, Slack, Discord etc. or launching with our built-in Webchat widget.

1. Webchat: Everything you need to know about our built-in Webchat widget like customizing the CSS, rendering custom widgets (like calendar pickers, file upload, and more), or customizing where and when it appears on your website.
2. Text-based interfaces: Code examples that we've built using the Dialog API to connect to interfaces like WhatsApp and Slack. You can clone these from GitHub and spin them up to connect your agent to these channels.
3. Voice-based interfaces: The same as above, but for Voice based interfaces.
4. Custom Interfaces: Code examples that we've built connecting Voiceflow to non-traditional interfaces like Unity to make AI generated game characters, Minecraft to create a co-pilot and more! Just to show how creative you can get with where you deploy your agent.

Improving Agents

Once you've got your MVP of your agent in production, it's time to start levelling up by creating more advanced use cases and expanding the functionality of what your agent can do.

1. Agent Enhancement Cookbook: Examples of how you can start connecting with custom LLMs, do advanced RAG filtering, increase speed with prompt compression and more. (these come straight from improvements we make to our own internal agents).
2. Leveraging RAG: Advanced ways to use the Knowledge base to have much more control over the accuracy of your AI answers.

Now, head on over to our Quick Start Guide to learn how to get started!