Quant-FX

Systematic strategies for gold and major FX

Each strategy is a fixed set of rules, written as code and tested on years of market data before it places a trade. The same process runs for our own master account and for every company that licenses the engine.

See the track record

What we trade

  • XAUUSDGold against the US dollarCore
  • EURUSDEuro against the US dollar
  • GBPUSDSterling against the US dollar
  • USDJPYUS dollar against the Japanese yen

Liquid instruments with tight spreads, so trading costs are predictable.

How signals are built

A signal is a list of conditions in code. The engine sends an order only when every condition holds, so nobody makes a judgement call on the day.

Trend

MA, ADX

Moving averages give the direction. ADX measures whether the trend is strong enough to trade.

Momentum

RSI, MACD

RSI and MACD have to confirm momentum behind the move before an entry is allowed.

Volatility

ATR

Average True Range filters entries by current volatility.

News blackout

Calendar

No trading from 30 minutes before to 30 minutes after major releases such as NFP, FOMC and CPI.

Every strategy is backtested in parallel, with spreads, commissions and slippage modelled

Ideas are prototyped in Python on AWS; the ones worth keeping are rewritten in C++20 for production.

Before a strategy goes live it passes walk-forward testing, out-of-sample validation and Monte Carlo stress tests on at least three years of data, across several market regimes.

Execution

What the engine does once a signal fires.

Entry
Limit orders at predetermined levels. No market orders during volatility spikes.
Exit
Every order goes in with its stop-loss and take-profit attached. Take profit sits at two to three times the amount at risk.
Sessions
London and New York only. No overnight holds on days with high-impact news.
Monitoring
Prometheus and Grafana track execution quality, latency, P&L and drawdown for every live strategy.

Risk limits

These are the defaults for a licensed deployment. They are set per client before a strategy goes live, and the engine enforces them. Our own master strategy account is a research environment and runs with wider limits. Its month-by-month history, including the losing month, is on the track record page.

Stop-loss per trade
Of account equity, on every trade.
1%
Open positions
At most, at any one time.
3
Position sizing
Quarter-Kelly sizing, with open risk across all positions capped at 3% of equity.
3%
Daily loss stop
Of equity. Once the day's losses reach it, the engine stops trading until the next day.
5%
Drawdown halt
Total drawdown of equity at which the engine halts trading.
10%

Built for licensing

Companies license the engine and the strategies that run on it. A company that wants the whole thing run for it can have us operate the strategies, copy engines and monitoring as white-label infrastructure.

The engine copies each trade from our master account into the client's own MT5 account. The client's broker executes it and holds the money.

How the licensing model works

Licensees remain responsible for their own regulatory obligations in their jurisdictions. See the Risk Disclosure.

Common questions

Yes. The engine places every order without manual input. We watch the systems daily, but we do not override signals.

Parameters are fitted on one stretch of data and then judged on a later stretch the fit never saw, which is what walk-forward testing does. A strategy that only works on the data it was fitted to does not go live.

Every fill is logged. We compare the expected price with the price we actually got and review the difference every month.

No. We never hold, manage or withdraw client funds. End clients keep their money in their own MT5 accounts at their own regulated brokers.

Yes. We build bespoke strategies and copy engines, and we can run them for you as white-label infrastructure. We supply the technology and do not give investment advice.

Commission a strategy

Tell us what you want to automate and we will send back a scoped plan.