-
AWS Architecture
-
Login Page
-
Sign up Page
-
Sign up Page
-
Recipe Page
-
Welcome Page
Inspiration
Our initial inspiration came from throwing out so much food from our fridge after going out for a long weekend. This made us realize that we could've made some food from the ingredients we had and saved a lot more food.
What it does
Our app can scan a person's grocery bill and extract the items that they purchased. We will have the expiry date of the products, and the app will suggest the user to use the products that are gonna be expired soon. Our app uses Generative-AI, with the help of AWS Bedrock leveraging LLM's, and also using Retrieval Augmented generation (RAG).
How we built it
A strong backend built with Python and Flask with MongoDB as our database and AWS for our LLM, and a smooth and reactive frontend built in react-native!
Challenges we ran into
Some significant challenges over the past 24 hours were
- the steep learning curve for react native for mobile development
- integrating aws bedrock LLM in addition to Dynamo DB for conversation memory
Accomplishments that we're proud of
- Creating the front-end, back-end, and linking both of them together.
- Implementing RAG and Generative-AI responses on custom knowledge base using AWS Bedrock.
What we learned
We learned a lot about AWS primarily. It was a new technology for all of us to use and it definitely paid off. The use RAG really cut down our recipe generation time. Further, We also learned about coding mobile apps in React Native which was definitely a huge step up from our previous experiences
What's next for WasteNet
The future functionalities that we're aiming to have for WasteNet, are including a chatbot for better user experience, in the long term, this app will be a part of large grocery stores, where AI can instruct the employees, when should the items be kept out to achieve minimal food wastage.
Log in or sign up for Devpost to join the conversation.