Cloud-Native AI built RESTful application launched via ECS Fargate

Project Overview

I recently designed and implemented a production-grade RESTful API service using Spring Boot and AWS cloud infrastructure, leveraging AI-assisted development practices. This microservice manages authors and their books, demonstrating modern cloud-native application development practices, Infrastructure as Code (IaC) principles, and effective AI tool utilization.

AWS Architecture of the RESTful Springboot Application

Technical Highlights

Backend Development

Developed a Spring Boot 3.x REST API using Java 17
Implemented JPA/Hibernate for database operations
Utilized Lombok for reduced boilerplate code
Built with Maven for dependency management and building
Containerized using Docker with multi-stage builds
Automated setup and troubleshooting via Terraform and Bash script

Cloud Infrastructure

AWS Services Utilized:

ECS (Elastic Container Service) with Fargate for serverless container orchestration
RDS (Relational Database Service) for MySQL database management
Application Load Balancer for traffic distribution
ECR (Elastic Container Registry) for container image management
CloudWatch for monitoring and logging
VPC with public and private subnets across multiple AZs
S3 Bucket for storing Terraform state file and DynamoDB for storing Terraform Lock file

$ ./scripts/setup-aws.sh prod
Setting up AWS infrastructure for environment: prod
Creating S3 bucket: terraform-state-author-books-prod
{                                                                                                                                                                                                           
    "Location": "http://terraform-state-author-books-prod.s3.amazonaws.com/"
}

Waiting for bucket to be created...
Enabling bucket versioning...
Enabling bucket encryption...
Blocking public access...
Creating DynamoDB table: terraform-state-lock-prod
{                                                                                                                                                                                                           
    "Table": {
        "AttributeDefinitions": [
            {
                "AttributeName": "LockID",
                "AttributeType": "S"
            }
        ],
        "TableName": "terraform-state-lock-prod",
        "KeySchema": [
            {
                "AttributeName": "LockID",
                "KeyType": "HASH"
            }
        ],
        "TableStatus": "ACTIVE",
        "CreationDateTime": "2024-11-10T11:34:09.441000-06:00",
        "ProvisionedThroughput": {
            "NumberOfDecreasesToday": 0,
            "ReadCapacityUnits": 1,
            "WriteCapacityUnits": 1
        },
        "TableSizeBytes": 0,
        "ItemCount": 0,
        "TableArn": "arn:aws:dynamodb:us-west-2:954976283848:table/terraform-state-lock-prod",
        "TableId": "752b33fb-4b73-4748-8bfb-2f6364016171",
        "DeletionProtectionEnabled": false
    }
}

DynamoDB table terraform-state-lock-prod already exists
Creating backend configuration...
AWS infrastructure setup complete!
Created/Updated:
- S3 Bucket: terraform-state-author-books-prod
- DynamoDB Table: terraform-state-lock-prod
- Backend Config: /c/Users/mtand/Projects/Dev/Java/springboot/authors-api-deployed-in-aws/terraform/environments/prod/backend.tfvars

Dev and Prod S3 buckets for managing Terraform state file

Dev and Prod DynamoDB tables for managing Terraform Lock file

Infrastructure as Code

Complete infrastructure defined using Terraform
Modular architecture with separate modules for
– Networking (VPC, subnets, route tables)
– Security (IAM roles, security groups)
– Container orchestration (ECS, ALB)
– Database (RDS)

Output of AWS resources created via Terraform

AI-Assisted Development

AI Prompt Engineering:

Leveraged Claude AI for architecture design and best practices
Utilized AI-assisted code generation and optimization
Implemented AI-driven infrastructure design patterns
Employed systematic prompt engineering for complex technical solutions
Used AI tools for code review and improvement suggestions

Architecture Highlights

High Availability:

Multi-AZ deployment with redundancy

As instance becomes unhealthy ECS Fargate automatically replace the instance with new instance

Security:

Network segregation with public and private subnets
Least-privilege IAM roles
Security groups for granular access control

Scalability:

Auto-scaling capabilities with ECS
Load balancing across multiple availability zones

Maintainability:

Infrastructure as Code and bash script for reproducible deployments
Containerization for consistent environments
Modular Terraform configuration for easier maintenance

Key Features

RESTful CRUD operations for authors and books
Relational data model with JPA/Hibernate
Automated infrastructure deployment
Production-ready configuration
Comprehensive error handling
Cloud-native design principles

Technical Skills Demonstrated

Development:

Java 17
Spring Boot 3.x
Docker
Terraform
REST API Design
Database Design

Cloud & Infrastructure:

AWS Services (ECS, RDS, ECR, VPC, IAM)
Infrastructure as Code
Microservices Architecture
Cloud Architecture
Security Best Practices

ECS Cluster with two running task launched via Fargate

Application image that was pushed to ECR

Created tables and inserted initial data into RDS database instance was SQL client

AI & Automation:

AI Prompt Engineering
Large Language Model Integration
AI-Assisted Code Development
AI-Driven Architecture Design
Systematic Prompt Construction
AI Tool Optimization

Development Practices

Infrastructure as Code (IaC)
Containerization
CI/CD readiness
Cloud-native design
Security-first approach
High availability patterns
Scalable architecture
AI-augmented development workflow

AI Integration Highlights

Prompt Engineering Excellence:

Crafted precise, contextual prompts for optimal AI assistance
Developed structured approaches to complex technical problems
Leveraged AI for architecture validation and improvement
Used AI tools for code optimization and best practices
Implemented AI-assisted documentation generation

Development Efficiency:

Reduced development time through effective AI tool usage
Improved code quality with AI-driven suggestions
Enhanced architecture decisions with AI-assisted analysis
Streamlined documentation process using AI tools
Accelerated problem-solving through AI collaboration

This project demonstrates proficiency in modern cloud application development, from backend programming to infrastructure automation, while showcasing the effective use of AI tools and prompt engineering for enhanced development efficiency. The combination of traditional development skills with AI-assisted methodologies represents a forward-thinking approach to software development.

Impact and Results

Reduced development time by leveraging AI-assisted coding
Improved code quality through AI-driven reviews and suggestions
Enhanced architecture design through AI-assisted planning
Accelerated infrastructure deployment using AI-optimized templates
Streamlined development workflow with AI tool integration

This hybrid approach of traditional software development skills combined with AI prompt engineering capabilities represents the future of software development, where AI tools are effectively leveraged to enhance productivity and quality while maintaining robust engineering practices.

Successful GET Request returns all books and author details

Successful POST Requests to add an author

Newly created Author added to the author table in the backend MySQL DB

Destroyed all the resources created on AWS via Terraform

I posted the project code to Github, you can find it here : https://github.com/aggrey-aggrey/springboot-restapi-built-with-claudeai-running-on-aws-ecs-fargate