MultiCloud, DevOps & AI eCommerce Platform

A multi-cloud powered eCommerce platform leveraging AWS, Google Cloud, and Azure, featuring automated CI/CD pipelines, Kubernetes orchestration, AI-driven recommendations, customer support, and Azure Text Analytics with Google BigQuery Integration.

---
Source Code available in the repository
Github handle: swetapati22
Project: CloudMart - MultiCloud, DevOps & AI eCommerce Platform
---

📂 Source Code: GitHub Repository
🎥 Demo Video: Watch Demo

Overview

CloudMart is an AI-powered eCommerce platform designed to operate seamlessly across AWS, Google Cloud, and Azure. This project demonstrates expertise in multi-cloud infrastructure, CI/CD automation, Kubernetes orchestration, and AI-powered integrations leveraging Amazon Bedrock, OpenAI Assistants, and Azure Text Analytics. Additionally, Google BigQuery Integration for Order Tracking & Analytics.

It covers in:

  • Automated Infrastructure Deployment using Terraform
  • Containerized App Deployment with Docker & Kubernetes
  • CI/CD Pipeline using AWS CodePipeline & CodeBuild
  • AI-Powered Product Recommendations with Amazon Bedrock
  • AI Chatbot for Customer Support powered by OpenAI
  • Order Tracking & Analytics using Google BigQuery
  • Sentiment Analysis with Azure Text Analytics

This project automates the entire cloud deployment lifecycle—from infrastructure provisioning using Terraform to end-to-end CI/CD deployment and AI-driven analytics.

Tech Stack

  • Cloud Providers: AWS, Google Cloud, Azure
  • Infrastructure as Code: Terraform
  • CI/CD Automation: AWS CodePipeline, AWS CodeBuild
  • Containerization & Orchestration: Docker, Kubernetes (EKS)
  • AI & ML Services: Amazon Bedrock, OpenAI, Azure Text Analytics
  • Data Processing & Storage: AWS DynamoDB, Google BigQuery

Deployment Steps

Each task is detailed in individual GitHub README files (link can be found within each task description below):

1️⃣ Provision AWS Infrastructure using Terraform

📄 See Detailed Steps

  • Deploy EC2 instances, IAM roles, and DynamoDB using Terraform.
  • Initialize Terraform and apply configuration: 
    cd terraform-project
terraform init
terraform apply

2️⃣ Build & Deploy CloudMart Application

📄 See Detailed Steps

  • Dockerize Backend & Frontend: 
    docker build -t cloudmart-backend application_source_code/backend/
docker build -t cloudmart-frontend application_source_code/frontend/
  • Push images to AWS ECR: 
    docker push <ECR_REPO_URI>/cloudmart-backend:latest
docker push <ECR_REPO_URI>/cloudmart-frontend:latest

3️⃣ Deploy Kubernetes Cluster on AWS EKS

📄 See Detailed Steps

  • Create an EKS Cluster: 
    eksctl create cluster --name cloudmart --region us-east-1 --nodes 1
aws eks update-kubeconfig --name cloudmart
  • Deploy the backend & frontend using Kubernetes manifests: 
    kubectl apply -f application_source_code/backend/cloudmart-backend.yaml
kubectl apply -f application_source_code/frontend/cloudmart-frontend.yaml

4️⃣ Set Up CI/CD Pipeline

📄 See Detailed Steps

  • Configure AWS CodePipeline to automate builds & deployments.
  • Add buildspec.yml to repositories for CI/CD automation.
  • Push changes to GitHub to trigger deployment.

5️⃣ Integrate AI & Data Processing

📄 See Detailed Steps

Amazon Bedrock for AI-Driven Product Recommendations

  • Deploy AWS Lambda function: application_source_code/backend/src/lambda/list_products.zip
  • Configure Amazon Bedrock Agent via Terraform in terraform-project/main.tf

OpenAI Assistant for AI-Powered Customer Support

  • Store API Key in application_source_code/backend/.env
  • Modify Kubernetes deployment file application_source_code/backend/cloudmart-backend.yaml to integrate the assistant.

Google Cloud BigQuery for Order Tracking & Data Analytics

  • Set up a BigQuery Dataset and create an orders table.
  • Deploy AWS Lambda: application_source_code/backend/src/lambda/addToBigQuery.zip

Azure Text Analytics for Sentiment Analysis

  • Update application_source_code/backend/cloudmart-backend.yaml with Azure API credentials.

Major Takeaways

  • Cloud-Native eCommerce: Fully automated deployment of a multi-cloud AI-powered eCommerce system.
  • Multi-Cloud Deployment: Seamless integration with AWS, Google Cloud, and Azure.
  • End-to-End CI/CD: Automated application builds & deployments via AWS CodePipeline.
  • AI-Powered Recommendations: Intelligent customer assistance using Amazon Bedrock for Product Recommendation Agent & OpenAI for Customer Support Agent.
  • Scalable Kubernetes Orchestration: Running backend & frontend on AWS EKS with Terraform.
  • Data Analytics Integration: Leveraging Google BigQuery & Azure Text Analytics for business insights.