Building a gas stations chatbot with AWS Lex, Lambda and HERE services:

In this workshop, we will deploy a small chatbot with AWS Lex and HERE services to help people find the gas stations near a specified location. We will explore the combination of AWS services, HERE Location Services and how they can be stitched together to create the beginnings of a powerful and scalable application. To complete this workshop, you will need: A personal AWS Account, basic knowledge of the AWS , an account with HERE and some basic understanding of node/Javascript.


  • HERE Developer Portal credentials.
  • Knowledge on AWS Lex and Lambda
alt text
Figure 1. testing the chatbot

Getting HERE credentials from HERE developer portal

To get the credential, head over to Click on the sign-up to get yourself registered.

Figure 2. AWS Market place

In the next page you will find a registration form, fill it up and get yourself signed up for Freemium plan.

PS:You don't need credit card for the subscription

Figure 3. AWS HERE over view page

In the next screen, you will find an optional phone number field, fill it if you want to get updates about HERE products and services. Now, agree to the terms and conditions and click on the Start Coding button to continue.

Figure 4. AWS HERE over view page

Now that you have successfully subscribed, click on the Generate App button to get the credentials.

Finally, you will get the codes as shown below

Figure 5. AWS HERE Sign-up page

Step 1: Deploy Lex bot

Login to your AWS console, search for Amazon Lex from the console and click on it. Now, click on the "create" button to get started and then on the next screen, click "Custom Bot" and fill in the details as shown in the screenshot below.

alt text
Figure 6. Step 1: Deploy Lex bot

Once you fill in the all the required details click on "Create" button to navigate to the next screen.

alt text
Figure 7. Create Intent

On the next screen you will see "Create Intent".

An Intent is a objective or action the user wants to achieve. Think of an intent as the specific reason a user would want to send a message to your bot.

Lets create an intent and call it "gasStationNearBy".

alt text
Figure 8. Create Intent

Once you have done that, you will be greeted by another screen with a many input options. Lets focus on the Slots section first.

alt text
Figure 9. Slots

You can think of Slots as the Variables of Lex. If someone was ordering a pizza, the Slots to include would be things like {SIZE}, {PIZZA_TYPE} and {DELIVERY_ADDRESS}.

There is a lot to discuss about slots and different slot types. Essentailly there are two types of slot available custom slots and builtin slots. For our chatbot we will be creating the custom slot. I encourage you to see more information about slots on the Amazon Docs.

For our chatbot, we will have one slot. "{gpsCodes}".

To create a custom slot, go ahead and add new slot type from the left hand side menu. You will find a dialog box which will have create slot type (As shown in the screenshot below).

alt text
Figure 10. Creating Custom Slots

Once you click on Create Slot Type you will be greeted with options to fill in. See the below screenshot to configure the options and go head and then click on the Add slot to intent button.

alt text
Figure 11. Creating Custom Slots

Once the slot is successfully saved, you can find the gpsCodes slot under the "Custom Slot Types" as shown in the below screenshot.

alt text
Figure 12. selecting the custom slot to the intent

Now that we have added the custom slot, let us go ahead and setup the slot for the Lex intent. In the following screenshot, we have to specify a name for our custom slot, from the slot type dropdown please select "gpsCodes" under the Custom slot types and then give a prompt like, "which gps coordinates?" Click on the "add" button under the settings to apply the slot to the current intent. Finally, click on the Save Intent button located on bottom of the page.

alt text
Figure 13. applying the slot to the intent

We need to add in variations of that utterance you can think of,but keep it relatively short to 2–3 different variations. Lets add the following utterances:

  • gas stations near {gpsCodes}
  • petrol stations near {gpsCodes}
  • fuel stations near {gpsCodes}
alt text
Figure 14. Adding uttetances

Next expand the Fullfillment options.

Here you have the option to either send the intent to a Lambda function or return parameters to client. For now, lets click Return Parameters to Client and click save.

We’re going to check on our bot before moving over to Amazon Lambda and creating the business logic of our chatbot. Scroll down and save your intent.

alt text
Figure 15. Adding Fulfillment

And now for a bit of fun! On the top right of the dashboard, click Build. This will initialize your chatbot and allow you to begin testing it. A chat widget should pop up on the right side of the screen. Try it out!

alt text Our Lex chatbot is almost ready to go. The next step is to attach a Lambda function to it.

The Lambda will simply receive the slots and their values and then return the in a way that Lex understands.

Step 2: Deploy a Lambda Function

In your AWS console, go to Amazon Lambda and select "Build a Function from Scratch". In the next screen, you will have the option of building your custom function or selecting a blueprint. The blueprints are great examples and I highly recommend checking them out to enhance your knowledge.

For our purposes, "Select Author From Scratch" and fill out the form with your function name, runtime environment (Node.js 10.x) and select the create function button to create the Lambda function.

alt text
Figure 16. creating lambda function

On this next Screen you will be given some boilerplate code for your lambda function, the designer tool, and some dashboard items for testing and monitoring your function. The following is the lambda function that we wanted to create:

PS: Don't forget to add the HERE credentials in the API Request to "".

'use strict';
const axios = require('axios')     
function close(sessionAttributes, fulfillmentState, message) {
    return {
        dialogAction: {
            type: 'Close',

// --------------- Events -----------------------

async function dispatch(intentRequest, callback) {
    console.log(`request received for userId=${intentRequest.userId}, intentName=${}`);
    const sessionAttributes = intentRequest.sessionAttributes;
    const slots = intentRequest.currentIntent.slots;
    const city =;

    const apiKey = process.env.apiKey;


          const gpsCodes = slots.gpsCodes;
    const codes = gpsCodes.split(',');
       // API to get the list of gas stations
              let gasStation = axios.get(`${codes[0].trim()},${codes[1].trim()}&cat=petrol-station&apiKey=${apiKey}`, {}).then((data)=>{

                let gasStationNames=>{

                    return ` *${it.title}* is at _${it.vicinity.replace("<br/>"," ")}_`+"\n"
                const reducer = (accumulator, currentValue) => accumulator + currentValue;

                callback(close(sessionAttributes, 'Fulfilled',
                {'contentType': 'PlainText', 'content': `Okay, Here are the list of gas stations ${"\n"}${gasStationNames.reduce(reducer)}`}));



// --------------- Main handler -----------------------

// Route the incoming request based on intent.
// The JSON body of the request is provided in the event slot.
exports.handler = (event, context, callback) => {
    try {
            (response) => {
                callback(null, response);
    } catch (err) {

The first time you hit Test, you will be asked to supply a JSON snippet describing what the test input should be.

  "messageVersion": "1.0",
  "invocationSource": "FulfillmentCodeHook",
  "userId": "user-1",
  "sessionAttributes": {},
  "bot": {
    "name": "HerePlacesBot",
    "alias": "$LATEST",
    "version": "$LATEST"
  "outputDialogMode": "Text",
  "currentIntent": {
    "name": "HerePlacesBot",
    "slots": {
      "gpsCodes": "18.000055,79.588165"
    "confirmationStatus": "None"

Here is the Lambda Deployment Package upload into the AWS Lambda function code dashboard as shown in the below screenshot.

alt text
Figure 17. creating lambda function

Now, select the Upload a zip file option from the dropdown to upload the downloaded zip file.

alt text
Figure 18. creating lambda function

Now, you see the lambda code in the Function Code section. Grab your APP_KEY from the HERE Developer Portal to setup the environment variables for our Lambda. alt text Once you have the APP_KEY, go ahead and setup the environment variables from the Environment Variables section.

Now that we have created out Lambda function, it is time to integrate it with Amazon Lex. Head over to the Lex dashboard and select the chatbot that we have already created to integrate the Lambda function.

alt text
Figure 19. creating lambda function

Go the chatbot settings and under the Fullfillment and select AWS Lambda function then choose the "getGasStationsNearBy" function. It will ask for the permissions, go ahead and click on OK

alt text
Figure 20. creating lambda function

Finally, click on the Save Intent button towards the bottom of the page to save all the settings.

Step 3: Testing the chatbot

In order to test our chatbot, click on the build button on the top right of the dashboard. It will take a while to build and once its done, you should be able to see chatbot section in the right side of the screen.

alt text
Figure 21. testing the chatbot

Go ahead and test it.


In this work shop we have build a chatbot powered by Lex, Lambda and HERE servies

alt text
Figure 22. testing the chatbot

results matching ""

    No results matching ""