AWS Builder Center

UseItUp: Building an AI-Powered Platform to Reduce Food Waste with AWS

UseItUp is a smart food management platform that helps people track their food, identify items nearing expiry, and discover recipes using ingredients they already have. In this blog, we explain how we built and deployed UseItUp using AWS services such as Lambda, API Gateway, DynamoDB, S3, CloudFront, Bedrock, Textract, SNS, and EventBridge.

Software Developer Intern

UseItUp: Building an AI-Powered Food Waste Reduction Platform with AWS

Food waste is a common problem in households. We often buy groceries, forget about them, and realize they have expired only when it is too late.
UseItUp was built to solve this simple but important problem: helping people understand what food they already have, what is close to expiring, and what they can cook with it.

What is UseItUp?

UseItUp is a smart food management platform that helps users:
  • Keep track of food items at home
  • Identify food that is close to expiring
  • Add food using bills or fridge photos
  • Get recipe suggestions based on available ingredients
  • Track nutrition information
  • Reduce unnecessary food waste
The main idea is simple:
Before buying more food or throwing food away, use what you already have.

The Problem

People regularly buy groceries but may forget what they have at home. Some food gets pushed to the back of the fridge and eventually expires.
This creates two problems:
  1. Food gets wasted
  1. Money is wasted on food that was never used
We wanted to build a system that could help users make better use of the food they already have without making food management complicated.

How UseItUp Works

The user can add their food items to the system using different methods.
For example, they can upload a grocery bill or take a photo of their fridge.
The system processes the information and adds the relevant items to the user's inventory.
The application then helps the user keep track of the items and their expiry information.
When users want to cook, they can ask the system for recipe suggestions based on the ingredients they already have.

Using AWS to Build the Application

We used AWS to build and deploy the application as a serverless system.

AWS Services Used

Amazon S3
We use S3 for storing uploaded files such as bills and fridge photos. We also use S3 to host the frontend files.
Amazon CloudFront
CloudFront delivers the frontend application to users through a public web URL.
AWS Lambda
The backend logic is divided into Lambda functions. This allows different features such as inventory, recipes, nutrition, uploads, and expiry checks to run independently without managing servers.
Amazon API Gateway
API Gateway connects the frontend with our Lambda backend and provides API endpoints for the application.
Amazon DynamoDB
DynamoDB stores information such as food inventory, user profiles, and other application data.
Amazon Bedrock
We use Amazon Bedrock with Amazon Nova 2 Lite to generate recipe suggestions based on the user's available ingredients.
Amazon Textract
Textract is used to extract information from uploaded grocery bills, helping reduce the amount of manual data entry required from users.
Amazon EventBridge
EventBridge is used to trigger scheduled expiry checks.
Amazon SNS
SNS is used as part of the notification system for expiry-related alerts.
AWS IAM
IAM manages permissions between the different AWS services and Lambda functions.
AWS SAM
AWS SAM was used to define, build, and deploy the serverless AWS infrastructure.

Technology Stack

For the application itself, we used:
  • HTML
  • CSS
  • JavaScript
  • Vite
  • Python
  • AWS SAM
For the cloud infrastructure, we used:
  • Amazon S3
  • Amazon CloudFront
  • AWS Lambda
  • Amazon API Gateway
  • Amazon DynamoDB
  • Amazon Bedrock
  • Amazon Textract
  • Amazon EventBridge
  • Amazon SNS
  • AWS IAM

Why We Used a Serverless Architecture

One of our goals was to avoid managing traditional servers.
With AWS Lambda and API Gateway, different parts of the application can run independently. DynamoDB provides the database layer, while S3 handles file storage.
This made it possible to build the application using separate services for separate tasks while keeping the architecture relatively simple.

AI in UseItUp

AI is mainly used for recipe generation.
Instead of simply showing a fixed list of recipes, the application can use the ingredients available in the user's inventory to generate recipe ideas.
For example, if a user has vegetables, rice, and a few other ingredients available, the application can use those ingredients to suggest what the user could prepare.
This makes the AI useful as part of the food management process rather than being added only as a separate chatbot feature.

What We Learned

Building UseItUp helped us understand how different AWS services can work together in a real application.
Some of the important things we worked with were:
  • Serverless backend development
  • API Gateway and Lambda
  • DynamoDB data storage
  • S3 file uploads
  • CloudFront deployment
  • AI integration using Amazon Bedrock
  • Document processing with Textract
  • Scheduled tasks using EventBridge
  • AWS IAM permissions
  • Infrastructure deployment using AWS SAM

Conclusion

UseItUp started with a simple observation: people often waste food because they forget about it or don't know how to use it.
By combining food tracking, expiry awareness, document processing, and AI-powered recipe suggestions, we built a platform that helps users make better use of the food they already have.
The project also gave us practical experience in building and deploying a complete application using AWS services.
Deployed Link -  https://d13eb3sn240lr1.cloudfront.net/  Try it !
Any opinions in this article are those of the individual author and may not reflect the opinions of AWS.
Enjoyed reading this content? Let the author know!

Your likes, comments, shares, and saves help creators reach more builders.

Loading recommendations

Loading article