Inspiration
With the holiday season rapidly approaching, it's impossible to keep track of everyone that needs a present. Grey Elephant uses AI to help you buy gifts that people actually want. Send your friends a link to sign in with their facebook, and you will get tailored gift ideas for them.
What it does
Searches online product catalogs to discover the best products for your friends. Using natural language processing to match product categories to interests.
How we built it
It's build on the Flask python micro-framework. We also pulled in various apis from facebook and ebay for product matching. The ML portion is done with sklearn and bs4 for webscraping. It was deployed on google-cloud's app engine, using a cloud sql instance for user data. HTML was styled with bootstrap 4.
Challenges we ran into
Poorly documented api's, difficult deployment, and lack of time were bothersome. App engine works great when you don't have alot of dependencies like we do.
Accomplishments that we're proud of
We made a pretty good looking website. We also were able to integrate the facebook login api and an email api to the project. We got the machine learning code working separate from the website.
What we learned
Programming in flask, web api calls, how web cookies work, how facebook authenticates people, how to do natrual language processing for a real app.
What's next for Grey Elephant
Getting the website fully functional, being part of amazon associates, more catalog api's, better handling of user data, more security, faster deployment pipeline.
Log in or sign up for Devpost to join the conversation.