ZenomTrader

The most insane thing i built with AI agents: A full hedge fund in a single file. 7 independent strategies. 2,000 lines of code. 5 symbols. One Expert Advisor. Each strategy has its own trade engine, its own risk parameters, its own magic numbers. They never touch each other. Why one file? Because that's how you control what actually matters: max drawdown across all strategies, overlap prevention so two algos don't stack the same directional bet, and anti-hedging logic for prop firm compliance. Nobody does this. Sharpe ratio above 2. Eight years of data. $100k starting balance. Claude Code launched MetaTrader, configured the parameters, ran the backtest, and switched to the equity curve. I didn't touch anything. Fully autonomous. I shared the exact logic and code structure inside the Discord. This is one of the most ambitious projects i've built so far:not just a strategy, a portfolio engine.

i just finished the biggest update to my personal Meta EA. Added two new strategies to the portfolio. Both completely uncorrelated with the existing ones. The equity curve with all strategies running together is smoother than any individual system on its own. This is what nobody understands about portfolio trading. You're not looking for the best strategy. You're looking for the strategy that improves the portfolio when combined with everything else. A system with negative expectancy in isolation can make the whole portfolio better if it's negatively correlated with the rest. All built with Claude Code agents. i described the logic, they wrote the NinjaScript, compiled it, debugged it, backtested it across years of data. i sat there watching the equity curve form in real time. insane. Every time i think i've hit the ceiling, the agents do something that surprises me again. One file. All strategies running simultaneously with overlap prevention and anti-hedging logic for prop firm compliance. Nobody does this because nobody knows how. it's actually harder than building separate EAs: you need shared risk management across all strategies, position sizing that accounts for correlation, magic number isolation. Still getting DMs asking me how to add a stop loss to a single NAS100 scalper. The gap is real.

32% win rate. 810 trades. Still profitable with FTMO I have an average profit of 2,090 and an average loss of 787. The reward to risk ratio is 2.65. The expectancy per trade is 143 Profitable both under regular market conditions and under the prop firm business model, my return on investment on FTMO is 10.8x If your mentor is laughing at a 30% win rate, why is their return on investment 5x at best? The math is simple: You don't need to be right most of the time, you just need to make more when you're right My track record is pretty messy before i leveled up on building new systems. I was only trading TSLA, taking survivorship bias to the extreme with a single instrument Now i'm trading a well diversified portfolio of strategies across multiple instruments and timeframes Most of the negative profit was in BTCUSD. It took me a few months to realize that crypto is untradeable with prop firms, the spreads and swap fees destroy any edge you might have Don't listen to the people telling you that the win rate matters, win rate is the biggest scam in trading. With 30% win rate you can still outshine someone with 95% win rate The industry is obsessed with win rate because losing 7 out of 10 trades feels bad. But feelings are not a trading strategy

People lose their minds when i tell them my win rate is around 30%. They think profitable trading means being right most of the time. it doesn't. It means making more when you're right than you lose when you're wrong. My average winner is multiple times my average loser. The expectancy is positive. The equity curve goes up over hundreds of trades. That's all that matters. i've seen traders with 85% win rates blow their accounts because their one losing trade wiped out months of tiny gains. Meanwhile i'm sitting here with consistent returns and a Sharpe that most quant funds would kill for. Win rate is the biggest scam the industry ever sold. They push it because 70% green trades looks good on a screenshot. Feelings are not a trading strategy. If your mentor is bragging about win rate ask them to show you the equity curve. Not a screenshot of one good week. Not a demo account. Years of live data across multiple instruments. i'll wait. crypto was a disaster by the way. The spreads and swap fees on prop firms make it basically untradeable. Took me months to figure that out. expensive lesson but at least i stopped bleeding on BTCUSD.

The problem with having a community is that you can't scale yourself. You answer the same questions hundreds of times. So i built something nobody in this industry has ever built: a full digital Doppelganger of myself. An autonomous AI agent trained on every message i've ever sent. 3 years of conversations, my tweets, my replies, my exact words. It runs 24/7 on my 80$/month VIP Discord. When someone opens a support ticket, my doppelganger answers exactly as if it was me. Same directness, same knowledge. It scraped thousands of my messages, every strategy post, every tutorial. If i already answered a question 6 months ago, it channels my exact words. without even thinking. it's actually better than me: it speaks any language. It never sleeps. it never forgets. When it doesn't know, it escalates to the real me. when i step in, it goes silent. The agent has boundaries. The autonomous hedge fund in a single file got surpassed. The most Promethean thing i've done. I pretty much automated my own presence and cloned my personality into a machine. The gap between people using AI agents and everyone else is becoming insane. If you can't see where this is going, you're already behind. Nobody has done anything like this, and it only gets better: every new message i send, it learns.

My doppelganger just handled a support thread that would have taken me 30 minutes to write out. It took seconds. It searched through years of my messages, found the exact conversation where i explained the same concept, and channeled my exact words. The member didn't even realize it wasn't me. He thanked me for the fast response. i was asleep. This is what scaling yourself actually looks like. Not hiring people. Not recording videos that get outdated in a month. An autonomous agent trained on everything you've ever written, running around the clock on the Discord. It handles support tickets. Explains strategy logic. Walks people through backtest setup. In any language. when it doesn't know something it escalates to me and goes silent. i sometimes go read the conversations it handled and i genuinely can't tell the difference between its answers and mine. The directness. The way it explains things. insane. I pretty much cloned my own presence. One person running everything. Zero employees. The doppelganger is better than a support team because it has my entire knowledge base and it never forgets anything i said 2 years ago. The most Promethean thing i've done and it only gets better. Every new message i send becomes training data. Still can't believe nobody else in this space has built anything like this. The gap between people using autonomous agents and everyone else is becoming an abyss.

I started with $1,000 to $2,000 in prop firm fees and compounded it into $200,000 in net payouts over three years. Sharpe ratio of 6+. CAGR approximately 300% per year. max drawdown around 15%. Most hedge funds never touch that. The secret? I stopped chasing the holy grail strategy and started doing what Jim Simons did at Renaissance Technologies. He wasn't running one perfect system. He was running tens of thousands of mediocre strategies with one thing in common: negative correlation. Like Simons said: "Negative correlation and as many systems as possible." I applied the same playbook to prop firm trading. Instead of optimizing one strategy, i run a portfolio of 7-10 systems that don't move together. When gold crashes, my mean reversion prints. When volatility spikes, my breakout systems win. When the market chops, the range-bound systems make money. PhDs managing billions still can't hit a Sharpe of 1. They laughed when i told them mine is 4+. But they're stuck optimizing for capacity and following rules that Jim Simons himself ignored. The lower the capacity, the higher the edge. I'm trading intraday patterns that billion-dollar funds can't touch because their size won't allow it. Here's the full breakdown:

The math behind prop firm compounding is insane when you actually run it. You start with a funded account. Pass the challenge. Get your first payout. Use that payout to fund more challenges. Each passed challenge gives you another account generating returns independently. i went from a couple thousand in fees to consistent monthly payouts across multiple accounts. Each one running a portfolio of uncorrelated strategies. If one account fails a rule, the others keep printing. The overall portfolio doesn't care. The whole secret is the same thing Jim Simons figured out decades ago. Run as many uncorrelated systems as possible. The individual systems don't need to be amazing. The portfolio needs to be amazing. Negative correlation is the real alpha. Most traders are still trying to find the one perfect strategy. They'll spend years optimizing a single system on a single instrument. Meanwhile i'm deploying new strategies with autonomous agents every week and the diversification keeps the max drawdown controlled. PhDs managing billions can't hit a Sharpe of 1. They're stuck optimizing for capacity. i'm trading intraday patterns that billion-dollar funds can't touch because their size won't allow it. i should probably track exactly when the compounding curve went exponential. it felt like nothing was happening for the first year and then it just broke through. Classic compounding. No team. No fund managers. No analysts. Just Claude Code agents and prop firm accounts. The lower the capacity, the higher the edge.

The S&P 500 is down 2.7% from all time highs. RIAG is up 9% YTD. One person. Zero employees. The #1 sharpe ratio fund in Romania, run entirely by autonomous AI agents. 672,000 $ in investor capital. 1.65 sharpe verified over 2 years. On friday we captured the entire Gold move at all time highs. No human touched a single trade. The agents identified the setup, executed, and locked profits while i was asleep. Most fund managers have 20+ analysts and developers. They spend millions per year on infrastructure. And most of them cant maintain a sharpe above 1.0 for a full year. I'm doing this alone from a small apartment in Romania with Claude Code and NinjaTrader 8. From strategy idea to live deployment in hours, not months. That's the real edge. Not the strategies themselves, but the speed at which i can build, test, and kill them. This is only the beginning.

Markets are choppy. Most funds are flat or negative this month. RIAG is green. One person. Zero employees. autonomous agents handling everything from strategy development to trade execution. Most fund managers would kill for a Sharpe above 1.0 sustained over multiple years. They have teams of analysts. They spend millions on infrastructure. They have compliance departments and risk committees and weekly meetings about meetings. i'm running this from my apartment with Claude Code and NinjaTrader 8. The agents build new strategies. The agents run backtests. The agents deploy to live. i make the final call and go to sleep. Last week the portfolio captured a move while i was offline. No human involved. The system identified the setup, sized the position, executed, and locked profits. i woke up and checked the equity curve. insane. The real edge isn't the strategies. it never was. It's the speed. From idea to live deployment in hours. If a strategy stops working i kill it and deploy a replacement before the drawdown matters. Try doing that with a team of 20 and a 6 month development cycle. i was actually looking at some hedge fund performance reports the other day. The fees they charge for returns that don't beat my worst month. it's almost funny. The gap keeps widening. This is still the beginning.