FiveM development that holds up at 700 players.
Custom scripts, custom interfaces, and the optimization work most devs skip. Six years building for GTA RP servers, full-stack, Lua through React.
- 6 yrs
- building FiveM
- 475
- resources audited in one server
- ~2,250
- redundant DB writes/min removed
- 343K
- views on my FiveM tutorials
What I do, and what I don't
I'd rather point you somewhere else on day one than take work I'm not the right person for.
I build
- Custom scripts and systems, built to your spec
- Custom NUI: HTML, CSS, and React interfaces
- Performance audits and server optimization
- Bug fixing on scripts you already own
- Reworking open-source resources into something yours
I don't
- Mapping and MLOs
- Clothing and EUP
- Server hosting and management
- Streaming and content for your server
Your mapping and creative team keeps doing what they do. I handle the code.
What hiring me actually looks like
A server that had been timing out for two months. Everyone, including the host, was chasing the wrong cause. This is the real sequence, from one week's work.
- symptom
Two mass timeouts in one night.
A 700-player ESX server hit 100% svmain at 1:00 AM and again at 3:00 AM, dropping everyone both times. The team had been chasing it for two months.
- theory
Slow database queries.
Everyone agreed, including the host. The database had already been migrated to external infrastructure on the strength of it. It was the one thing that felt actionable.
- measured
The database was idle.
Read-only inspection of the live MariaDB: 99.9998% buffer pool hit rate, one thread running, and every "slow" query resolving as a one-row primary key lookup.
- cause
The database was never the bottleneck.
svmain was blocked, and a blocked server thread inflates what oxmysql reports as query time. The metric everyone trusted was measuring the symptom, not the cause.
- shipped
Five commits, same day.
Batched an odometer script writing ~2,250 unbatched rows a minute. Deleted a server-wide broadcast firing on every money mutation that nothing listened to. Pulled ALTER TABLE out of two hot paths. Added yields to the paycheck sweep.
Kept anonymous because it's a client's server, not because the numbers are soft. Every figure above is in the commit log, and I'll walk you through the full report on a call.
Three ways to start
Every one of them starts with a conversation and a written scope. You approve the price before I write any code.
Server audit
Fixed price, quoted
I profile the whole resource tree, find what's actually eating your server thread, and hand back a prioritized report. Remediation quoted separately, or I ship the fixes myself.
Best if the server stutters, times out, or nobody can tell you why.
Retainer
From $500 / month
Ongoing development. Custom scripts, bug fixes, and optimization on a rolling basis, tracked on a shared board so you always know what's in flight.
Best if you're actively building and want a developer on call.
One-off build
Quoted per project
A single script or interface, scoped and priced up front. You approve the spec before I write code, and the price does not move after that.
Best if you know exactly what you want built.
Tell me what you're building
The more detail you give me, the more accurate the quote. I read every one of these myself.