Projects

Building a Serverless Recipe-Sharing Application

Serverless
Lambda
API Gateway
DynamoDB
Cognito

Deploying a fully serverless recipe web application using AWS managed services and Terraform.

Introduction

In this project I designed and deployed a fully serverless recipe-sharing web application on AWS.

The objective was to understand:

  • serverless architectures
  • frontend and backend separation
  • authentication and authorization
  • infrastructure automation
  • scalable application design

This project is especially useful for:

  • full-stack cloud applications
  • serverless development
  • secure API design
  • infrastructure automation

Services Used

The project uses the following AWS services:

  • Amazon S3
  • Amazon CloudFront
  • Amazon API Gateway
  • AWS Lambda
  • Amazon DynamoDB
  • Amazon Cognito
  • IAM
  • Terraform

Architecture

CloudFront
      ↓
S3 Frontend
      ↓
API Gateway
      ↓
Lambda
      ↓
DynamoDB

Authentication
      ↓
Cognito

Configuration

Frontend Deployment

Frontend configuration included:

  • S3 static website hosting
  • assets storage
  • CloudFront distribution
  • Origin Access Control (OAC)

Best practices:

  • restrict direct S3 access
  • cache frontend content
  • separate frontend infrastructure

Database Layer

DynamoDB configuration included:

  • recipes table
  • scalable billing mode
  • Lambda integration
  • NoSQL storage

Authentication

Authentication was implemented using Amazon Cognito.

Configuration included:

  • User Pool
  • user authorization
  • admin creation
  • email verification

Important note:

Cognito sends temporary credentials by email

Backend Services

The backend was implemented using Lambda and API Gateway.

Implemented endpoints:

  • Authentication
  • Health Check
  • Get Recipes
  • Create Recipe
  • Delete Recipe
  • Like Recipe

Frontend Configuration

After infrastructure deployment, frontend values must be updated.

Required files:

aws-exports.ts
configs.tsx

Values retrieved from Terraform:

  • AWS Region
  • Cognito User Pool ID
  • Cognito Client ID
  • API Gateway URL

Frontend Build

Build process:

npm install
npm run build

Upload:

content from 'dist/'
↓
S3 Bucket

Lessons Learned

This project helped reinforce:

  • serverless design
  • authentication workflows
  • frontend deployment
  • backend orchestration
  • infrastructure automation

AWS Well-Architected Best Practices

Operational Excellence

  • Infrastructure provisioned using Terraform.
  • Repeatable and automated deployments.
  • Clear separation between frontend and backend components.

Security

  • Authentication and authorization implemented using Amazon Cognito.
  • Least-privilege access control through IAM roles and policies.
  • Direct access to backend resources restricted through API Gateway.
  • CloudFront Origin Access Control (OAC) used to protect S3 content.

Reliability

  • Fully managed AWS services reduce operational overhead.
  • DynamoDB provides built-in high availability.
  • Serverless architecture removes dependency on individual servers.

Performance Efficiency

  • CloudFront improves global content delivery performance.
  • Lambda automatically scales based on application demand.
  • DynamoDB provides low-latency access to application data.

Cost Optimization

  • Pay-per-use serverless services minimize infrastructure costs.
  • No EC2 instances required.
  • DynamoDB and Lambda scale according to actual usage.

Sustainability

  • Serverless services automatically adjust resource consumption.
  • Managed AWS services reduce infrastructure waste.

Project Results

Key outcomes:

  • Successfully deployed a fully serverless recipe-sharing web application on AWS.
  • Implemented secure user authentication using Amazon Cognito.
  • Built a scalable backend using Lambda and API Gateway.
  • Stored and managed application data using DynamoDB.
  • Delivered the frontend globally through CloudFront and S3.
  • Automated the infrastructure deployment using Terraform.

References

Project Code

You can find the Terraform code used in this project in the following GitHub directory.

View Project Directory

Comments

Loading comments...
Loading...