AI Agents

A conversation connects a user with an agent in order to achieve a specific goal. In Moveo, the agent role can be played by a human, as traditionally, or by an AI agent, which is a more advanced version of the human agent. These AI Agents are the core of Moveo's conversational logic. They have several features and components that you can use to customize their behavior and achieve your business goals.

note

AI Agents were formerly called Brains. If you previously used the platform under this terminology, check out this section to learn more about the new AI Agent structure.

How do AI Agents respond?

Moveo AI Agents have two different ways to respond to a user's message: dialogs and collections. Although they serve different purposes, they can also work together to create more personalized and dynamic conversations. For example, you can use live instruction variables from dialogs to enhance collection-based answers on the fly.

Dialogs

Dialogs

are structured workflows that the AI Agent will follow during a conversation if a specific trigger is activated. They are composed of blocks connected to each other, each serving a specific function.

Dialogs are particularly useful when you need to:

• Connect to external services and APIs through webhooks.
• Implement complex logic trees and conditional flows.
• Automate specific tasks like changing a password, validating identity, or managing subscriptions.

Knowledge

Knowledge is the combination of a collection

, and custom guidelines defined for the AI Agent. Collections are the source of information for the AI Agent, and custom guidelines the instructions on how to use that information and handle the conversation.

To manage this behavior, AI Agents have a dedicated knowledge

section. This is where you configure how the AI Agent uses the LLM. You define both the static content it can access—such as collections—and the guidelines that instruct the AI how to respond using that information. Think of it as telling the agent what it knows and how it should use that knowledge to reply appropriately.

AI Agent types

AI Agents can be of different types, each of them trained and optimized for a specific use case. Some of them have predefined guidelines that can be used to enhance the agent's behavior.

tip

Complete the predefined guidelines to get the best out of your AI Agent.

• Early Engagement
  Engage customers early to build relationships, answer questions, and offer info before they commit.

• Product Adoption
  Guide customers through easy product adoption and setup for a quick start.

• Upsell
  Encourage exploring higher-tier options with relevant upgrades and benefits.

• Debt Collection
  Help manage overdue payments with flexible options and reminders for smooth settlements.

• Customer Support
  Provide customer support, addressing inquiries, troubleshooting, and FAQs.