How LineOnMe AI works
The parts of LineOnMe AI: provider and model, AI employees, knowledge, channels, conversations, clients, analytics and the activity log.
Updated October 9, 2026
LineOnMe AI is made of several parts that work together. Knowing how they connect helps you set the system up faster and understand why the AI replied the way it did.
AI provider and model
This is your employees’ “brain”. You connect a provider with your own API key and choose a model. One provider can serve several employees, and for important scenarios you can set a backup model.
AI employees
Each employee is a role with instructions, knowledge, channels and permissions. The employee decides what to reply to a client and proposes actions: some it performs on its own, and important ones it sends to you for approval.
Knowledge
Documents, price lists, rules and short facts about your business. When replying, the employee finds the relevant passages in them and relies on them.
Channels and Conversations
Messages from your website chat, messengers and email arrive in Conversations. One client’s messages are gathered into a single thread, even if they write to different channels.
Clients and Products and services
Clients appear automatically from conversations. The catalog of products and services gives employees exact prices and availability, and sales are recorded right in the conversation.
Analytics and the activity log
Analytics shows how fast and how well you reply, and the activity log shows what each AI employee did and who approved it.