SmartPath-AI

Cloud-native AI application deployed on AWS EKS using Terraform, Github Actions CI/CD, ArgoCD GitOps, and observability via Prometheus, Grafana.

Table of Contents

• Overview
• Architecture
• Technology Stack
• Repository Structure
• Prerequisites
• Getting Started
• CI/CD Pipeline
• Acknowledgments

Overview

SmartPath finder AI is a production-grade cloud-native platform running on AWS EKS and managed through Terraform, GitHub Actions, and ArgoCD GitOps. The project demonstrates modern DevOps practices including Infrastructure as Code (IaC), GitOps workflows, automated CI/CD, secrets management, Kubernetes-native observability, and scalable application deployment.

Key features:

• Infrastructure provisioning with Terraform
• GitOps deployment model using ArgoCD App-of-Apps
• Automated CI/CD pipelines with GitHub Actions
• Kubernetes-native observability with Prometheus and Grafana
• External Secrets integration with AWS Secret Manager
• Persistent storage through AWS EBS CSI Driver
• Production-ready EKS architecture

Architecture

The platform consists of three independent repositories

Infrastructure Repository:

• Responsible for provisioning AWS infrastructure
• VPC networking
• Amazon EKS cluster
• IAM Policies
• IRSA Roles
• EBS CSI integration
• Storage Classes

GitOps Repository:

• Acts as the single source of truth for Kubernetes resources
• ArgoCD Applications
• App-of-Apps orchestration
• Infrastructure components
• Monitoring stack
• MongoDB deployment
• Application deployment

SmartPath-AI_Application:

• AI application source code
• Docker configuration
• GitHub Actions workflows

---

3-Tier application architecture Diagram

AWS infrastructure architecture Diagram

Technology Stack

Category	Technologies
Cloud	AWS
Infrastructure as Code	Terraform
Containerization	Docker, EKS
GitOps	ArgoCD
CI/CD	Github Actions
Version Control	GitHub
Database	MongoDB
Storage	AWS EBS CSI Driver
Secrets Management	External Secrets Operator
Security	IAM, IRSA, RBAC
Networking	Ingress NGINX
Application	Java Script, HTML, CSS, Express.js, Bootstrap, EJS
Observability	Prometheus, Grafana

Repository Structure

├── Application
│   ├── Dockerfile
│   ├── README.md
│   ├── app
│   │   ├── db.js
│   │   ├── index.html
│   │   ├── index.js
│   │   ├── node_modules
│   │   ├── package-lock.json
│   │   ├── package.json
│   │   ├── public
│   │   └── views
│   ├── docker-compose.yaml
│   ├── mongo-init.js
│   └── nginx
│       ├── errors.txt
│       ├── logs.txt
│       └── nginx.conf
├── GitOps
│   ├── app-servicemonitor.yaml
│   ├── apps
│   │   ├── app-of-apps.yaml
│   │   ├── application.yaml
│   │   ├── database.yaml
│   │   └── infra.yaml
│   ├── database
│   │   ├── dbvalues.yaml
│   │   └── mongodb
│   ├── infra
│   │   ├── ingress-nginx
│   │   ├── ingress.yaml
│   │   ├── sc-ebs-csi-gp3.yaml
│   │   └── secretstore.yaml
│   ├── mongodb-values.yaml
│   ├── monitoring
│   ├── nginx-controller-values.yaml
│   ├── path-finder-ai
│   │   ├── Chart.yaml
│   │   ├── charts
│   │   ├── templates
│   │   └── values.yaml
│   └── prometheus-stack-values.yaml
├── Infrastructure
│   ├── README.md
│   ├── iam-policy.json
│   ├── iam_policy.tf
│   ├── main.tf
│   ├── modules
│   │   ├── ebs-csi-storageclass
│   │   └── network
│   ├── providers.tf
│   ├── terraform.tfvars
│   └── variables.tf

Getting Started

📊Observability

Monitoring

• Prometheus: Metrics collection from application and infrastructure
• Grafana: Custom dashboards for application performance and node monitoring
• Endpoints: /metrics for application monitoring

Dashboards

• Node monitoring and resource utilization
• Application-specific metrics and performance

Acknowledgments

• Infrastructure Repository
• GitOps Repository
• Application Repository

Note

The project is in progress, more to be updated.