What I take on
Prop Firm Platform.
Built by me, start to deploy
A funded-trader firm that can sell a challenge, judge it against live equity and pay the trader who passes.
Why this gets built
Most firms rent this whole thing. The rules you can write and the data you can read are whatever the vendor decided, and when a trader disputes a breach nobody on your side can open the record and answer. The engine underneath is where that gets settled. Manual review does not scale, and neither does the first engine that replaces it: mine stalled around 500K evaluations an hour on database and Redis round trips. That ceiling arrives quietly. The queue lags, a breach is called late, and the trader has already traded past it.
What I build
- Challenge phases, profit targets, drawdown, daily loss, consistency and minimum days held as rows rather than branches in code. Changing a rule is a form, not a deployment.
- The evaluation engine underneath, reading live equity off the trading server instead of a nightly export, so a breach is called when it happens rather than the next morning.
- Every verdict stored with the numbers it was made from, so a disputed breach is answered in a minute.
- A scheduler that can re-run one account, a whole firm, or a backfill across a date range without a maintenance window.
- Trader portal showing live equity, the rules running against it, and the breach that ended a phase with the numbers behind it.
- Staff desk that can open an account, stop one, move a trader between stages or re-open an evaluation that closed wrongly, without a ticket to a vendor.
- Challenge checkout, then the activation path a trader walks after a pass.
- One database you own, so every breach, payout and account change is queryable by your own people.
How long it takes
Around eight weeks to a firm that can sell a challenge, judge it and pay a trader. The engine on its own, running rules against a platform you already have, is about four.
The 30-day launch on agency pages assumes a team. Alone I would rather cut scope than the calendar: four weeks gets a working challenge with a trader portal, and the payout and identity rails take the weeks after that. A second pass for throughput is a separate job with its own estimate. The Rust rebuild at FTMO was only worth starting once the ceiling was real, and how long one runs depends on what the hot loop turns out to be.
What ships
- Trader portal
- Staff desk
- Challenge checkout and activation
- The evaluation engine and its rule schema
- Backfill and replay tooling, and verdict history with the inputs behind each one
- Load numbers measured on your own data rather than a demo set
- Admin screens for rule changes
- Payout request queue
- Deployment, runbook and a handover session
Tech stack
Next step
Let’s talk, 30 minutesNeed this built?
Tell me what you’re running and where it breaks. Half an hour is usually enough for me to say what it takes and what I’d do first.