AI chatbot call center: Demo call center (production-ready, part 2)

Workflow Name: ☎️ Demo Call Center

Template was created in n8n v1.90.2

Skill Level: High

Categories: n8n, Chatbot

Stacks

• Execute Sub-workflow Trigger node
• Chat Trigger node
• Redis node
• Postgres node
• AI Agent node
• If node, Switch node, Code node, Edit Fields (Set)

Prerequisite

• Execute Sub-workflow Trigger: Telegram Call In Workflow (or your own node)
• Sub-workflow: Taxi Service (or your own node)
• Sub-workflow: Taxi Booking Worker (or your own node)
• Sub-workflow: Demo Call Back (or your own node)

Production Features

• Scaling Design for n8n Queue mode in production environment
• Optional Rate Limit design to prevent overused
• Optional Long Terms Memory design
• Multi-Service design
• Testing Flow with or without dependance on other workflow.
• Error Management

What this workflow does?

This is a n8n Demo Call Center Workflow demo. It is the main entry node for a Multiple Services Chatbot. It will receive message from the Call In Workflow, and decide which service should be use, or which service provider should be process the selected result.

How it works

• The Flow Trigger node will wait for the message from the Call In Workflow or other Sub-workflow.
• When message is received, it will first check for the Rate Limit.
• If ok, load the Session Data from Cache.
• Next, check the current Session for the channel_no (default is chat).
• if channel_no is chat, continue to the AI Agent for chit-chat.
• if channel_no is taxi or others, pass to the Service Input (i.e. Taxi Service) or Service Worker (i.e. Taxi Booking Worker) to handle it directly.
• The AI Agent should decide which service (i.e. taxi) will be used at some point and update the channel_no in Session, and pass to the Service Node (i.e. Taxi Service) at once.
• In case of any error, reply the error in Call Back.

Set up instructions

1. Pull and Set up the required SQL from our Github repository.
2. Create you Redis credentials, refer to n8n integration documentation for more information.
3. Select your Credentials in Rate Limit, Session, Provider and New Session.
4. Create you Postgres credentials, refer to n8n integration documentation for more information.
5. Select your Credentials in Postgres Chat Memory, Load User Memory and Save User Memory.
6. Modify the AI Agent prompt to fit your need

How to adjust it to your needs

1. In Session, we have a timestamp fields which is created at the Input node. The usage of this is combined to use with the session id to create a unique session, since some media, such as Telegram, do not have a unique session along with the chat.
2. You can use any AI Model for the AI Agent node
3. Learn we use the prompt for the Load/Save User Memory on demand.
4. Include is our prompt for the taxi service. You can add more service similar to this.