Project · 0022023 — 2026

Private Online Platform

I handled infrastructure, backend systems, deployments, monitoring, and production issues for a large FiveM platform supporting real communities and daily player traffic. Most of the work was making systems reliable, scalable, and easy to operate long-term.

RoleLead dev & sysadmin
StackFiveM · NGINX · Node
Reach4M+ creators
Team6 engineers
01 / Context

Reliability under real traffic

The platform supported large public communities with hundreds of concurrent players, live events, voice systems, databases, web dashboards, and real-time game services running together.

My work focused on keeping everything stable during updates, debugging production issues quickly, and building systems that were easy to maintain over time instead of relying on temporary fixes.

Most of the infrastructure was self-managed across Linux and Windows hosts using NGINX, Node.js, Redis, MongoDB, PostgreSQL, Cloudflare, and internal tooling.

02 / Operations

Production operations

Most production issues came down to debugging networking, server performance, caching problems, database bottlenecks, deployment mistakes, or game resource instability.

I usually approached incidents by reproducing the problem first, narrowing down possible causes, validating fixes in development, and rolling changes out carefully to avoid impacting live players.

We also built internal documentation, deployment procedures, backups, and monitoring tools to reduce repeated issues over time. A lot of the work was making sure updates could be reverted quickly if something broke in production.

03 / Systems

Infrastructure stack

The platform was split into edge routing, backend services, game servers, and multiple database systems depending on workload requirements. Different parts of the stack prioritized low latency, persistence, or operational simplicity.

  • Edge & proxy

    NGINX reverse proxies handled routing, SSL termination, caching, static assets, and traffic forwarding across multiple internal services and game endpoints.

  • Application tier

    Node.js services powered APIs, dashboards, automation systems, WebSocket features, and internal tooling used by both players and staff.

  • Data tier

    MongoDB, PostgreSQL, MySQL, and Redis were used depending on the workload. Work included backups, indexing, migrations, replication planning, and performance debugging.

CLIENTFRONTEND
GATEWAYAPI
DATAPRISMA / POSTGRES
Fig. A: Edge → services → data (simplified)
04 / Scale

What the numbers looked like

4M+

Combined creator audience

100+

Peak concurrent players

20+

Production services managed

What I took from running it

Observability first

Good logs and monitoring save hours during incidents. Being able to quickly trace failures mattered more than adding more infrastructure.

Small deployments

Smaller, reversible updates made production safer and easier to debug compared to large releases.

Documentation matters

Internal docs, deployment notes, and runbooks made onboarding and maintenance significantly easier over time.