This project demonstrates an end-to-end DevOps workflow by deploying a simple AI-based text analysis application using modern cloud-native tools.
Users enter text in a web UI.
The application analyzes the sentiment using a lightweight AI service.
Flow: Frontend → Backend API → AI Service → Response
- Frontend: Static HTML + JS (served via Nginx)
- Backend: FastAPI service
- AI Service: FastAPI-based text sentiment analyzer
- Containerization: Docker
- CI/CD: GitHub Actions
- Image Registry: AWS ECR
- Orchestration: Kubernetes (planned)
- Packaging: Helm (planned)
- Infrastructure: Terraform (planned)
| Service | Port | Description |
|---|---|---|
| frontend | 80 | Web UI |
| backend | 8000 | API gateway |
| ai-service | 8001 | Text analysis engine |
- On every push to
main:- Build Docker images
- Push images to AWS ECR
All sensitive values (AWS credentials) are stored securely using GitHub Actions Secrets.
- Terraform-managed AWS infrastructure
- Kubernetes (EKS) deployment
- Helm charts for services
- Ingress with HTTPS
- Observability (Prometheus + Grafana)
Built by Henal Mehta as a hands-on DevOps learning project.