-
AI-DevOps-Assistant is an end-to-end cloud-native DevOps and AIOps platform that demonstrates how modern applications can be built, deployed, monitored and intelligently managed on AWS.
-
The project combines Kubernetes, GitOps, Infrastructure as Code, CI/CD automation, observability and Generative AI into a single production-inspired architecture.
-
The application follows GitOps principles where Git remains the single source of truth and ArgoCD continuously synchronizes the Kubernetes cluster with the latest repository changes.
-
Infrastructure provisioning is completely automated using Terraform while GitHub Actions builds, scans, packages and publishes every microservice automatically.
-
The deployed application runs on Amazon EKS and is continuously monitored using Prometheus, Grafana, Alertmanager and Amazon CloudWatch.
- Build a Cloud-Native Microservices Platform
- Implement Infrastructure as Code using Terraform
- Provision an Amazon EKS Cluster
- Deploy containerized workloads on Kubernetes
- Implement GitOps using ArgoCD
- Automate CI/CD using GitHub Actions
- Store Docker images in Amazon ECR
- Secure workloads using IAM, OIDC and IRSA
- Manage secrets using AWS Secrets Manager
- Monitor applications using Prometheus & Grafana
- Centralize logs using Fluent Bit & CloudWatch
- Integrate Amazon Bedrock Agents for AIOps
- Build an AI-powered DevOps Assistant
- Amazon VPC
- Amazon EKS
- Managed Node Groups
- Amazon ECR
- IAM Roles
- IAM OIDC Provider
- IAM Roles for Service Accounts (IRSA)
- AWS Secrets Manager
- GitHub Actions
- Automated Docker Builds
- Security Scanning
- Amazon ECR Push
- Automatic Image Versioning
- Kubernetes Manifest Updates
- ArgoCD
- Auto Sync
- Self Heal
- Pruning
- Drift Detection
- Desired State Management
- Deployments
- Services
- ConfigMaps
- Secrets
- StatefulSets
- Namespaces
- Service Accounts
- Rolling Updates
- Prometheus
- Grafana
- Alertmanager
- Fluent Bit
- Amazon CloudWatch
- Metrics Collection
- Log Aggregation
- Dashboard Visualization
- Amazon Bedrock Agent
- Lambda Action Groups
- Root Cause Analysis
- Log Investigation
- Metrics Analysis
- Service Health Monitoring
- AI Recommendations
Developer
β
βΌ
GitHub Repository
β
βΌ
GitHub Actions CI Pipeline
β
ββββββββββΊ Build Docker Images
ββββββββββΊ Security Scan
ββββββββββΊ Push Images to Amazon ECR
ββββββββββΊ Update Kubernetes Manifests
β
βΌ
GitHub Repository
β
βΌ
ArgoCD
β
βΌ
Amazon EKS Cluster
β
βββββββββββββββΌββββββββββββββ
βΌ βΌ βΌ
PostgreSQL Microservices Monitoring
β
βΌ
Amazon Bedrock Agent
β
βΌ
AI DevOps Assistant (KIRA)
| Layer | Technology |
|---|---|
| Frontend | React |
| Backend | FastAPI / Python |
| AI Assistant | Streamlit |
| AI Platform | Amazon Bedrock |
| AI Model | Qwen3 32B |
| AI Actions | AWS Lambda |
| Database | PostgreSQL |
| Containers | Docker |
| Local Development | Docker Compose |
| Container Registry | Amazon ECR |
| Kubernetes | Amazon EKS |
| Infrastructure | Terraform |
| CI | GitHub Actions |
| GitOps | ArgoCD |
| Monitoring | Prometheus |
| Dashboards | Grafana |
| Alerts | Alertmanager |
| Logging | Fluent Bit |
| Log Storage | Amazon CloudWatch |
| Secrets | AWS Secrets Manager |
| Authentication | IAM + OIDC + IRSA |
AI-DevOps-Assistant/
β
βββ .github/
β βββ workflows/
β βββ ci.yml
β
βββ Architecture/
β βββ architecture.png
β
β
βββ docs/
β
βββ gitops/
β βββ k8s/
β βββ manifests/
β βββ kustomization.yaml
β βββ argocd.yaml
β
βββ projects/
β
β βββ boutique-microservices/
β β
β βββ Infrastructure/
β β
β βββ aiops-assistant/
β
βββ README.md
β
βββ LICENSE
| Microservice | Responsibility |
|---|---|
| Frontend | React Web Application |
| Gateway API | Routes requests to backend services |
| Auth Service | Authentication & Authorization |
| User Service | User Management |
| Product Service | Product Catalog |
| Order Service | Order Processing |
| Orders Service | Order Management |
- User Information
- Product Catalog
- Orders
- Authentication Data
- Application Metadata
Push Code
β
βΌ
Checkout Repository
β
βΌ
Install Dependencies
β
βΌ
Run Lint/Test
β
βΌ
Build Docker Images
β
βΌ
Security Scan
β
βΌ
Push Images to Amazon ECR
β
βΌ
Update Kubernetes Manifests
β
βΌ
Commit Updated Image Tags
- frontend
- gateway
- auth
- user-service
- product-service
- order-service
- orders
GitHub
β
βΌ
ArgoCD
β
βΌ
Detect Changes
β
βΌ
Sync Cluster
β
βΌ
Deploy New Version
- Amazon VPC
- Public Subnets
- Route Tables
- Internet Gateway
- Amazon EKS Cluster
- Managed Node Group
- Amazon ECR Repositories
- IAM Roles
- IAM Policies
- IAM OIDC Provider
- Helm Releases
- ArgoCD
- Monitoring Stack
- IAM Roles
- IAM Policies
- IAM OIDC Provider
- IAM Roles for Service Accounts (IRSA)
- AWS Secrets Manager
Prometheus
- Collect Metrics
- Kubernetes Monitoring
- Service Metrics
Grafana
- Dashboards
- Visualization
- Performance Monitoring
Alertmanager
- Alert Rules
- Notifications
- Incident Alerts
Fluent Bit
- Log Collection
- Kubernetes Log Forwarding
CloudWatch
- Centralized Log Storage
- Log Search
- Operational Visibility
- Fetch Logs
- Fetch Metrics
- Check Service Health
Example:
Why is my Product Service unhealthy?
- Invokes Lambda
- Retrieves metrics
- Collects logs
- Checks service health
- Generates an AI response
-
The AI-DevOps-Assistant extends a traditional DevOps platform by integrating Amazon Bedrock Agents to automate operational tasks using Generative AI.
-
Instead of manually checking dashboards, logs, and Kubernetes resources, engineers can interact with an AI assistant using natural language.
-
The assistant can investigate production issues, analyze metrics, inspect logs, and perform service health checks through AWS Lambda Action Groups.
User
β
βΌ
Streamlit Web Interface
β
βΌ
Amazon Bedrock Agent (KIRA)
β
βββββββββββββΌβββββββββββββ
βΌ βΌ βΌ
Fetch Logs Fetch Metrics Health Check
β β β
βΌ βΌ βΌ
AWS Lambda AWS Lambda AWS Lambda
β β β
βββββββββββββΌβββββββββββββ
β
βΌ
Amazon CloudWatch / Kubernetes
β
βΌ
AI Generated Response
- Large Language Models
- AWS Lambda
- CloudWatch Logs
- Kubernetes Monitoring
- Natural Language Interaction
- Bedrock Agent
- Foundation Model
- Prompt Orchestration
- Agent Instructions
- Reading application logs
- Returning recent log entries
- Assisting log investigations
- CPU Usage
- Memory Usage
- Request Count
- Response Time
- Kubernetes Metrics
- Pod Status
- Deployment Status
- Service Health
- Cluster Health
- Application Availability
- AI Chat Interface
- Root Cause Analysis
- Metrics Inspection
- Log Analysis
- Health Checks
- Natural Language Queries

