Himanshu NagrathApr 27, 2026
Build an AI WhatsApp Receptionist for Appointments 🤖
35views
Build an AI WhatsApp Receptionist for Appointments 🤖
Himanshu Nagrath
Learn how to build an AI-powered receptionist for your dental clinic on WhatsApp. This tutorial covers setting up system prompts, configuring tools for saving data, and integrating calendar events to automate appointment bookings.
Build an AI WhatsApp Receptionist for Appointments 🤖
Himanshu Nagrath
Learn how to build an AI-powered receptionist for your dental clinic on WhatsApp. This tutorial covers setting up system prompts, configuring tools for saving data, and integrating calendar events to automate appointment bookings.
Chapters
00:00
Introduction to AI Agents
Overview of building an AI receptionist for a dental clinic on WhatsApp.
00:37
Understanding Agent Components
01:41
Generating System Prompts
03:00
Configuring Save Attributes
04:06
Calendar Integration Setup
05:30
Testing the Agent
Transcript
00:00
Hey there.
00:00
today I'm going to show you how you can very quickly build your own AI agent on WhatsApp.
00:05
Right.
00:05
So let's get started in the automation.
00:07
You can create a new automation.
00:13
I'm going to name this, AI receptionist.
00:18
Right?
00:19
So I'm designing an AI receptionist which will help me to book an appointment for a dental clinic.
00:24
Right?
00:25
So I've chosen a business number.
00:27
this is like a WhatsApp business number.
00:29
You can add your own number on the settings/WhatsApp.
00:32
I've already done that.
00:33
so inside agent.
00:36
So yeah, basically you can create multiple nodes, but for this one I'm gonna create an AI agent.
00:42
an AI agent has four different components prompt which, where you will define the system prompt.
00:47
This is you will elaborate what an AI agent is supposed to do.
00:50
Give it a personality.
00:52
Knowledge base is something where you can provide it, websites or PDFs.
00:57
Basically the knowledge base required answer the queries by the end customer.
01:02
Right.
01:02
But for this case we don't need a knowledge base.
01:05
Tools are something where we can add third party integration so that an AI agent can take actions.
01:11
Right.
01:11
And finally routes are basically the exit paths where you can define like when an appointment has been confirmed.
01:17
Then you can like send an email notification to yourself or assign a ticket to some other human, person.
01:24
Right.
01:24
So bunch of different kind of functionalities are available, right?
01:28
you can generate your own system prompt using AI as well.
01:30
So let me show you how to do that.
01:33
generate a system prompt for an AI receptionist working at a dental clinic.
01:37
your job is to help, patients book an appointment with the clinic.
01:40
please make sure to collect these important information such as email, date of appointment, time of appointment and the service they need.
01:48
the dental clinic provides services such general consultation, root canal, teeth whitening, teeth cleaning, and some of the services as well.
01:59
So, right, we can generate the prompt here.
02:03
So what I have done here, I have given instructions to an AI to generate a system prompt for an AI.
02:09
Right?
02:09
So, so this is the best way to generate system round because system run needs to be very elaborative.
02:14
And so in my case, like it has created with dental clinic a receptionist, what are the responsibilities, what information they need to collect?
02:22
Email address, appointment date, appointment time.
02:24
Right.
02:25
So all of these are already added, in the.
02:26
Now we can go to tools.
02:28
So we have two different kind of tools.
02:29
One is Save Attribute tool.
02:31
Save Attribute tool allows you to save important information inside your audience along with the contact.
02:36
Right.
02:37
So I'm gonna name it same save attributes.
02:39
use this tool to save appointment booking information.
02:46
date of appointment.
02:48
And then time of appointment.
02:54
email address of the patient.
02:59
And we also need service, right?
03:01
kind of service that the patient is looking for at the dental clinic.
03:06
Right.
03:06
So I have given all the instructions.
03:09
So this tool will be attached to the it will save all this important appointment information inside the audience.
03:15
Right.
03:16
the next thing that I need is to a calendar event.
03:19
So create calendar event.
03:28
Right.
03:29
So this part is a bit I've already done that for the demo.
03:33
yeah.
03:34
Right.
03:35
we have a calendar create calendar event.
03:37
This will create the booking using Cal.com APIs on my calendar.
03:41
Right.
03:42
So there are dynamic variables like start date this date time has to be in ISO so LLM will generate the start date time for the API request body.
03:51
I have added headers like CAL API version and authorization then rest of the things which are in blue color are coming from audience attributes.
03:58
Email, name, phone number, all of these are coming from audience attributes.
04:02
Start date, time is something which is dynamic in nature and LLM is going to generate that.
04:06
Right.
04:07
So these are all the information that I need.
04:09
So these are the tools that I've added.
04:11
Right.
04:12
So now let's get started and test this out.
04:16
So I'm gonna switch on this one.
04:21
Let's use.
04:27
It.
04:40
So let's start our AI agent now.
04:44
Hello, welcome to dental Clinic.
04:51
I want for the 26th of March, 11am.
04:57
What's her email address?
04:59
Email?
05:09
Your oneness is confirmed.
05:11
So let's check it out.
05:15
my email has been saved and these are the, information that I gave.
05:19
So root canal appointment date and appointment time has been saved.
05:25
So I can see all the live, transcription of the conversation here as well.
05:30
And yeah, that's all.
Comments
Comments
No comments yet