Kamlesh Prasad Production Portfolio & Cloud Infrastructure
High-availability web hosting on AWS EC2 with automated GitHub Actions push-to-deploy pipeline, Nginx reverse proxy, and Cloudflare edge CDN caching.
Executive Overview & Context
Engineered an automated cloud delivery architecture for a production client web application. The platform incorporates zero-downtime Nginx reloads upon release tag creation, TLS 1.3 cryptographic termination, automated Let's Encrypt certificate renewal, and enterprise email security records (SPF, DKIM, DMARC).
Project Objectives
1. Build a zero-downtime deployment pipeline triggered by Git tags. 2. Enforce edge security and DDoS mitigation via Cloudflare. 3. Implement automated TLS certificate lifecycles via Let's Encrypt and Certbot. 4. Protect enterprise email deliverability through SPF, DKIM, and DMARC policies.
Challenges & Pain Points
Replaced error-prone manual FTP/SSH uploads with a strict continuous delivery pipeline, eliminated customer-facing downtime during content deployments, and mitigated public cloud attack surfaces through rigorous security group controls.
Engineered Implementation
Configured an AWS EC2 instance running Ubuntu 24.04 LTS behind Cloudflare DNS and CDN caching. Connected GitHub Actions to trigger automated test suites, build minified production bundles, and execute zero-downtime deployment scripts over authenticated SSH keys.
Key Engineering Features
System Topology & Traffic Flow
Client Browser → Cloudflare Global Edge CDN → AWS EC2 Security Group (Ports 80/443) → Nginx Reverse Proxy with TLS termination → Node.js application process supervisor.
Cloud Delivery Architecture
End-to-end edge-to-compute traffic flow and automated CI/CD pipeline
Client Browser
Cloudflare Edge Network
Nginx Web Server & EC2
Vercel Edge Functions
Google Gemini AI Engine
GitHub Actions CI/CD
Health & Alert Automation
| # | From → To | Flow / Action | Protocol / Type |
|---|---|---|---|
| 1 | Client Browser→Cloudflare Edge Network | HTTPS Ingress | TLS 1.3 |
| 3 | Cloudflare Edge Network→Nginx Web Server & EC2 | Origin Request | HTTPS (Port 443) |
| 2 | Client Browser→Vercel Edge Functions | AI Chat Invocations | HTTPS / POST /api/chat |
| 4 | Vercel Edge Functions→Google Gemini AI Engine | Prompt Inference | REST API / TLS 1.3 |
| 5 | GitHub Actions CI/CD→Nginx Web Server & EC2 | SCP Dist & SSH Reload | SSH (Port 22) |
| A | Health & Alert Automation→Nginx Web Server & EC2 | Daily Health Probesecondary | HTTP Localhost |
Inspect Detailed Cloud & Security Architecture
Explore full multi-layer topology, live active ports, security controls, and verified repository evidence in the Architecture Explorer.
Implementation Details & Workflow
Provisioned AWS EC2 in region ap-south-1 with 20GB gp3 EBS storage. Configured systemd service descriptors and PM2 ecosystem configuration. Created a dedicated deploy user with restricted sudo privileges and public key authentication.
CI/CD & Automation Pipeline
GitHub Actions workflow runs on every push to main. It installs dependencies, runs linter and unit tests, executes production build, and executes an atomic blue/green or symlink reload script on the server.
Security & Hardening Controls
Configured UFW default-deny policy. Enforced Strict-Transport-Security (HSTS), X-Content-Type-Options: nosniff, and Content-Security-Policy headers in Nginx configuration. Implemented rate-limited SSH ingress.
Technologies & Cloud Services
Measurable Results
Achieved 100% automated release velocity with zero manual SSH commands required for deployments. Global time-to-first-byte (TTFB) reduced significantly via Cloudflare edge caching.
Ensuring zero dropped HTTP connections during Nginx reload was solved by utilizing systemctl reload nginx combined with PM2 graceful reload signals.
