AWS Builder Center

Weekend Annoying Task Challenge: AI Fridge Meal Planner

An AI-powered meal planning web application that creates a 7-day meal plan based on family size, cooking time, cuisine preferences, and ingredients available in the refrigerator.

Introduction

Planning dinner every day can be frustrating as a parent, especially when you already have ingredients in your refrigerator but don’t know what to cook.
For this Weekend Challenge, I built an AI-powered web application called AI Fridge Meal Planner. The application creates a personalized 7-day meal plan using Amazon Bedrock (Amazon Nova Lite) based on family size, available cooking time, preferred cuisine, and ingredients already in the refrigerator.
To build this project, I used several AWS serverless services, including Amazon S3 Static Website Hosting, Amazon API Gateway, AWS Lambda, Amazon Bedrock, Amazon CloudWatch, and AWS IAM.

The Problem

Many people spend time deciding what to cook each day.
Common challenges include:
  • Not knowing what to cook
  • Not knowing what ingredients are already in the refrigerator
  • Spending too much time planning meals
  • Tired of planning meals every day
I wanted to create a simple web application that makes meal planning easier and less stressful.

The Solution

The application asks users for:
  • Family size
  • Available cooking time
  • Preferred cuisine
  • Ingredients currently in the refrigerator
After the user submits the form:
  1. Amazon API Gateway sends the request to AWS Lambda.
  1. AWS Lambda creates a prompt using the user’s input.
  1. Amazon Bedrock (Amazon Nova Lite) generates a personalized 7-day meal plan.
  1. The meal plan is returned and displayed on the webpage.
Because the application uses AWS serverless services, there is no server to manage.

AWS Services Used

Amazon S3 Static Website Hosting

The frontend (HTML, CSS, and JavaScript) is hosted using Amazon S3 Static Website Hosting.
This provides a simple and cost-effective way to host a static web application.

Amazon API Gateway

Amazon API Gateway receives requests from the frontend and securely forwards them to AWS Lambda.

AWS Lambda

AWS Lambda processes the user’s input.
The function:
  • Receives the request
  • Creates the AI prompt
  • Invokes Amazon Bedrock
  • Returns the generated meal plan to the frontend
Because Lambda is serverless, there is no infrastructure to manage.

Amazon Bedrock

Amazon Bedrock powers the AI functionality.
The application uses the Amazon Nova Lite foundation model to generate personalized 7-day meal plans based on the user’s preferences and available ingredients.

Amazon CloudWatch

Amazon CloudWatch automatically stores Lambda logs.
During development, I used CloudWatch to troubleshoot errors and verify that the Lambda function was running correctly.

AWS IAM

AWS IAM securely manages permissions.
The Lambda execution role allows the function to invoke Amazon Bedrock and write logs to Amazon CloudWatch.

Application Architecture

The application follows a simple serverless architecture shown below.
Architecture diagram of the AI Fridge meal Planner application.

  1. The user opens the website hosted on Amazon S3.
  1. API Gateway forwards the request to AWS Lambda.
  1. Lambda creates a prompt and calls Amazon Bedrock.
  1. Amazon Bedrock generates the meal plan.
  1. Lambda returns the response to the website.

What I Learned

This project helped me better understand how AWS serverless services work together.
Before building this application, I understood each AWS service individually. This project helped me understand how Amazon S3, API Gateway, AWS Lambda, Amazon Bedrock, Amazon CloudWatch, and AWS IAM work together to build a complete AI application.
I also learned how to design effective prompts for generative AI and integrate Amazon Bedrock into a serverless application.

Challenges

CSS styling took much longer than I expected, and I spent a lot of time adjusting the user interface.
This was my first time participating in the Weekend Challenge, so I underestimated the amount of time needed to complete the project.
I ended up finishing close to the deadline, but I learned the importance of better time management. Next time, I will start earlier and leave more time for testing and improvements.

GitHub Repository

Live Demo

Conclusion

This was my first time participating in the Weekend Challenge. Completing and submitting this project increased my confidence in building AWS applications.
Building this project gave me hands-on experience with AWS serverless technologies and generative AI, which were both new to me.
Using Amazon Bedrock for the first time was especially exciting. I didn’t know this service existed before, and I was impressed by how easily it could generate personalized meal plans.
This project inspired me to continue learning more about Amazon Bedrock and other AWS AI services. 
I’m excited to build more projects and gain more hands-on experience with AWS.
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