Inspiration
I have previously tried to make something similar where a bot would use sms text to make appointments in Twilio. Seeing that there was a challenge about this made us interested.
What it does
It allows customers to check store information like website, hours, address, and phone number. It also allows users to check their appointments and repair orders scheduled. The store is a dealership.
How we built it
The store information and customer information was provided in JSON files. Using AWS LexV2, we able to create logic diagrams for a chat bot to speak fluently and take in orders. Using AWS Lambda, we were able to further customize our bot to use custom replies. The bot can handle wrong information and can ask again user for correct information.
Challenges we ran into
We got confused between Lex V1 and Lex V2 documentation. It took as a while to notice but how they accept lambdas are different. Learning Lex V2 documentation was also new along with the change of user interface and additional dialogs needed.
Accomplishments that we're proud of
We are proud that we got everything that we finished everything we set out to do. A proper bot assistant with a store info, client info, and appointment booking features. We also implemented emojies.
What we learned
We learned how to make a powerful smart assistant that is integrated with SMS text.
What's next for Dealership Smart Assistant
Add gifs and pictures as replies. Add support for multiple users and store locations. Use a Lambda function for initialization and validation and not just for fulfillment.
Credits
Yoel Robaina
Amazon Lex bot configuration
Python programming
Video editing
Mauricio Silva
Python programming
Debugging
Jonathan Montesinos
Research
Graphic design
Jordan Hickerson
Amazon Lex bot configuration
Narration


Log in or sign up for Devpost to join the conversation.