I’m building a Forex trading bot with almost no budget. Who’d like to work on it with me?
Posted by Naqi · Location-independent
Building a trading bot with almost zero starting cash, testing it on Forex until it hits a 90% accuracy target, and then turning it into an investment fund - has anyone tried a path like this?
Here are a few things I'm thinking about:
1. The Accuracy Trap: Chasing a 90% win-rate can actually be a bit of a trap in trading. Sometimes a strategy with a much lower win-rate makes way more money if its winning trades are significantly bigger than its losses. Managing drawdowns and the risk-to-reward ratio usually matters a lot more than just the raw accuracy percentage.
2. Keeping Costs Low: Coding and backtesting are basically free using tools like Python or MQL5, but you still run into minor costs for reliable historical data or a fast VPS to keep the bot running smoothly.
3. The Regulatory Wall: Moving from trading your own tiny account to actually taking outside money and running a fund is a massive jump. Dealing with legal compliance, licensing, and broker rules is usually the hardest bottleneck to clear.
I’d love to work with people who know this world: a developer, someone with good data, a tester, or anyone with ideas on where a project like this could go.
Building a trading bot with almost zero starting cash, testing it on Forex until it hits a 90% accuracy target
Why I’m posting this
I want to get feedback on my idea.
Open to
Partnership, Collaboration, Sale, Sponsorship
Free public Ideas
Helpful early warning: a 90% hit rate is the wrong target. Many bots that win 90% of trades still lose money, because a few large losses wipe out many small wins, and spreads plus commissions eat thin edges. Track net profit after costs and maximum drawdown instead, and test on data you didn't tune on. That's where most bots quietly fail.
A useful collaborator could challenge the strategy rather than help optimise it. Write down the rules and evaluation method first, then have someone independently reproduce the results, including unsuccessful versions. A transparent research record would give a future technical partner something concrete to assess.
Ideas are free public suggestions. Proposals are paid and private and are never shown here.