Internal tool on Private VPS Deployment Expert Needed

I have a stack of custom internal business tools that must go live on my own VPS and stay there with at least 99.9 % uptime. The suite covers:

• Data management
• Workflow automation
• Reporting & analytics
• AI prompting with RAG integration

All code is ready; what I need now is fault-tolerant deployment, solid security hardening, and round-the-clock monitoring.

The work in practical terms
– Provision and tune the Ubuntu-based VPS (firewall, SSH, swap, kernel parameters).
– Containerise the applications (Docker or Podman) and orchestrate them via Docker Compose or a lightweight Kubernetes flavour such as k3s—whichever you can keep lean and maintainable.
– Configure Nginx/Traefik reverse proxy with SSL (Let’s Encrypt) and automatic renewal.
– Set up a CI/CD pipeline (GitHub Actions or GitLab CI) so pushes to the main branch trigger blue-green deployments with zero downtime.
– Build observability: Prometheus + Grafana (or similar) for metrics, Loki/ELK for logs, plus actionable alerts to Slack/email.
– Automate nightly off-site backups of databases and persistent volumes.
– Produce concise run-books and hand-over documentation that let any future admin reproduce the environment.

Acceptance criteria
1. Fresh install script/provisioner completes without manual intervention and passes health checks.
2. Automated pipeline rolls out a trivial update in under five minutes with no user-visible interruption.
3. Synthetic uptime monitor shows ≥99.9 % over a seven-day burn-in period.
4. Security scan (OpenVAS/Clair/Trivy) reports no critical findings.

I’ve already ruled out canned hosting panels and one-click solutions, so hands-on experience with complex, from-scratch deployments is essential—ideally 10 + years in the trenches. If your response reads like it was written by a bot, it will be ignored. Show me a real-world example, mention the main tools you plan to use, and outline how you’ll meet the acceptance criteria.