One Person, a Dozen Production Services

Problem

Every case study on this site runs somewhere — and "somewhere" is a single self-managed server, not a managed platform abstracting the operational work away. Running a dozen-plus independent services reliably on one box, without a platform team, means the infrastructure design itself has to carry the reliability burden.

Action

• Designed and operate a Docker Compose-orchestrated stack running 12+ polyglot production services — Django apps, WordPress installs, Node.js services, Python-based Telegram bots, and n8n automation workflows — side by side on shared infrastructure.
• Run Nginx as the shared reverse proxy and TLS termination layer in front of all services, with Certbot handling automated certificate issuance and renewal across every domain and subdomain.
• Manage data persistence across three different database engines (PostgreSQL, MariaDB, SQLite) chosen per service based on what that service actually needs, rather than forcing one engine across everything.
• Structured deploys so individual services can be rebuilt and redeployed independently with zero downtime to the others — a config or code change to one bot or app doesn't risk an outage on an unrelated service sharing the box.

Result

A dozen-plus production services — spanning several languages and frameworks — running reliably on infrastructure one person designs, deploys, and maintains end to end. It's the same operational discipline (isolate blast radius, automate the repetitive parts, know exactly what's running where) that shows up in every client engagement on this site, just pointed at Ali's own stack instead of someone else's.