Excercise: Automate a Complex Workflow Using AI
Advanced AI-Driven Workflows for Complex Support Needs
In this video, we’ll create a fully automated AI workflow that collects customer data, sends it to your backend system via REST API, captures the response, and uses that information to dynamically guide the conversation—all without any manual intervention.
You’ll learn how to:
🤖 Detect customer intent and trigger the right workflow
📝 Collect and store customer data using conversation properties
🔗 Send data to your backend via REST API integration
📥 Capture and store API responses for intelligent decision-making
🔄 Guide conversations dynamically based on API results
🧪 Test and validate your automated flow to ensure it works flawlessly
✅ Example Scenario: Imagine a customer requests eligibility verification. Your AI instantly collects their info, checks with your backend, and provides a personalized response—all automatically!
📌 Perfect for:
Business owners, automation enthusiasts, and anyone looking to create powerful AI workflows that save time and scale operations.
Transcript
Auto-generated from the video captions; it may contain errors.
In this exercise, you will create a fully automated chat flow using AI. The goal is to collect data from a customer, send it to the back end via REST API, capture the response, and use that information to dynamically guide the next step in the conversation. Okay. Okay. Okay. Okay. Okay. By completing this, you will see how AI can seamlessly handle complex task without any manual intervention. Let's get started. First, create an intent. Define an intent that detects when a customer wants a trial extension. Second, you can add a mission to collect the data. Create a mission to collect the required customer information. Add conversation properties to store the data.
For example, account ID or email address. Then set up a REST API integration. Map the payload fields to the conversation properties like for example the account ID. Capture the API response. Create a new conversation property to for example LGPD status to store response from the API. Then map the API response field to this property. Now create another mission. Add a new mission to handle the next step based on the API response. Define the condition to check the eligibility status directly in the prompt and if eligible respond positively and share details. If not eligible offer alternative solution or next step. Okay. Run a test conversation to ensure the flow works as expected.
Check that the data is collected, sent to the back end and the assistant responds correctly based on the API response. By the end of this exercise, your AI assistant aura should recognize the intent of course collect the customer data and send it to a backend system via rest API. It will capture the response and store it in a property dynamically respond based on the response using a mission goal. This chat flow will fully automate the process from start to finish. Once you have completed the exercise, think about how you can use this structure for other complex workflows in your business.