We are big (ant)eaters and never like to let food go to waste. The food sold on the marketplace might not look the best, but that doesn't mean it doesn't taste the best. Don't judge a book by its cover!

What it does

This webapp allows you to buy food that is not appealing to the eye and consumers for a cheaper price that. Farmers and wholesalers that are not able to sell the food at normal price are able to add products webapp. Users can register and buy food from the home page using the Stripe api.

How we built it

We started by building the webapp locally using Flask and Python in VSCode. Once we reached a point where the webapp had unique menus, we registered a domain to and started working on deploying the webapp to GCP.

Challenges we ran into

Due to some GitHub push and pull errors early on when we were building the webapp locally with Flask, our modules would not call correctly in the .yaml file. We had to completely reconstruct the webapp from the ground up in Google's App Engine on the cloud to fix the import errors we were getting. We were unable to develop a dynamic way to add users to the webapp and connect them in a reasonable way to the shopping cart, so we allow the consumers to buy as a guest. We also had trouble implementing the machine learning code into the webapp after we trained the model to be 91% accurate theoretically.

Accomplishments that we're proud of

We are proud of making a webapp that can revolutionize the way we buy and sell food. The Lucky Harvest can increase the access to more affordable food and decrease the amount of food the population wastes. Actually getting a machine learning model deployed was also cool too!

What we learned

We learned Bootstrap, html design, Jinja2, Flask, Google Cloud AutoML Vision Object Detection, databases ,etc

What's next for The Lucky Harvest

We were not able to implement the machine learning directly into the webapp, but the next steps would be to allow the sellers to take a picture of the produce they want to sell and the webapp's inventory would automatically be updated with the correct food. This would make the time to list food on the website exponentially faster and show that sustainability can both be good for the environment and easy to maintain. We would also like to implement the Google Maps api to make the process of buying and selling less nuanced and be directed towards people that live near each other.

Share this project: