Plutux

AI trading-system builder

Write your trading rules in plain words. Get a system you can backtest.

Plutux is an AI trading-system builder: describe your entry and exit rules in a sentence, and it backtests them on real market history, paper-trades them live, and tells you the moment one fires.

Browse 50 documented systems
  • Rules about news, not just numbers
  • Real market history
  • Paper trade on live prices
  • Alerts the moment a rule fires
New systemAI
Illustrative preview. Real backtests run on actual market history inside Plutux.
Your system
BacktestSystemBuy & hold

System

Max drawdown

Win rate

Trades

How it works

Three steps. Plain words in, a tested system out.

  1. 1

    Write the rules the way you'd say them

    “Buy when RSI drops below 30 and volume beats its 20-day average.” Or “sell if the news around it turns negative.” Both are rules here — Plutux turns either sentence into cards you can read and edit, entry on one side, exit on the other.

    You see: your rules as a flow, not a formula.

  2. 2

    Backtest on real history

    Pick the stocks and the years. Every trade the rules would have made is listed with its reason, next to a buy-and-hold line for the same stocks.

    You see: an equity curve, drawdown, and a plain-language read of what worked.

  3. 3

    Paper trade it, get told when it fires

    Let the system watch your stocks on live prices. When a rule triggers, it emails you and shows up in your signal inbox — you decide, it keeps score.

    You see: a running account, every wake-up logged, and whether you followed your own plan.

What a backtest tells you

Not a score. A record of every decision the rules would have made.

A single number can come from luck. Plutux shows you the trades behind it, next to what buying and holding the same stocks would have made, so you judge the system rather than the outcome.

  • Headline numbers, then the honest ones

    Total return, max drawdown, win rate and closed trades up front. Annualized return, Sharpe, profit factor and exposure one click further, with a note on what each one can and cannot tell you.

  • Buy-and-hold on the same stocks

    The fair comparison: what the same money would have earned by buying on day one and never touching it. If the rules didn't beat doing nothing, you see it in one line.

  • Every trade with its evidence

    Entry, exit, return, and which rule fired, with what it saw at that moment — “close 182.3 above SMA50 179.8”, or the headline a news rule read. Replay the run bar by bar and watch the marks land on the chart.

  • News rules read only what was published by then

    A rule about the news or an earnings call is answered from the articles and transcripts that existed on that date, and nothing after it. Sources that only exist as a snapshot of today are refused rather than backfilled — that is the one shortcut that makes a bad system look brilliant.

  • Warnings a spreadsheet won't give you

    Under thirty trades, win rates are mostly noise, so it says so. Tune a run five times and it reminds you that the honest next test is live data no tuning has seen.

Paper trading and signals

The system watches the market so you don't have to.

Add the stocks you care about to a system's watchlist and go live. Plutux checks them against your rules six times a trading day, opens and closes paper positions by those rules on real prices, and tells you the moment something happens.

  • Email the moment a rule fires

    Before the open, at the open, and into the close. When a condition is met, or is within a percent of it, the alert lands in your inbox naming the rule that triggered.

  • A signal inbox that keeps score

    Every check is logged: what fired, on which stock, at what price, and what the paper account did about it. Record the trades you act on and the journal fills itself in.

  • Your rules vs. your hands

    Rule adherence is the share of your trades your system actually asked for. It's the number this product exists to raise, and it's visible from your first live week.

Six checks a day · US market, Eastern time

  1. 08:30Pre-open gap check
  2. 09:35Entry rule fired · AAPL
  3. 10:30Open + 60 min
  4. 13:00Midday check
  5. 15:45Close − 15 min
  6. 16:15Close + 15 min · verdict

Start from a documented system

Not sure where to start? Clone a system that already has a name.

Fifty well-known systems — Turtle rules, Darvas boxes, Donchian breakouts, Connors RSI(2) and more — taken apart rule by rule. Clone one, then change what doesn't fit you.

  • Breakout

    Donchian Channel Breakout

    Buy a new four-week high, sell a new four-week low, and do nothing else. The rule set behind the Turtles.

    BeginnerPosition (weeks to months)
  • Trend following

    Golden Cross / Death Cross

    Own the market while the 50-day average is above the 200-day, and stand aside while it is not.

    BeginnerPosition (weeks to months)
  • Mean reversion

    Double Seven Mean Reversion

    The shortest complete rule set here: one filter, one entry, one exit, and the only number in any of them is seven.

    BeginnerSwing (days to weeks)
  • Momentum

    Dual Momentum (GEM)

    One switch a month between stocks, bonds, and cash. Antonacci's whole book in a rule.

    BeginnerLong term (years)
  • Macro

    60/40 Balanced Portfolio

    Hold 60% stocks and 40% bonds, and put the weights back to target once a year.

    BeginnerLong term (years)
  • Value

    Dogs of the Dow

    Once a year, buy the ten highest-yielding stocks in the Dow, hold them untouched for twelve months, then re-rank.

    BeginnerLong term (years)

Who it's for

For investors who are ready to stop guessing.

Plutux helps you turn a simple investing idea into rules you can test, understand, and improve — even if you have never built a trading system before.

  • 01

    Start with what you already think

    You do not need a formula or a finished strategy. Describe an idea in your own words — buy after a pullback, stay out of a downtrend, sell when the story around a company turns — and Plutux turns it into clear rules you can read and edit.

  • 02

    See what would have happened

    Test those rules on real historical prices. You see when they worked, when they failed, and what the market was doing each time, instead of relying on instinct alone.

  • 03

    Build a process you can repeat

    Improve your rules as you learn, and you end up with a system you can follow and review — instead of deciding from scratch every time a stock moves.

Questions

Frequently asked

What is an AI trading-system builder?+

A tool that turns trading rules written in plain language into a system a computer can test and follow. In Plutux, you describe entry and exit conditions in a sentence; the AI converts them into structured rules, backtests them on real historical prices, paper-trades them on live data, and alerts you when a condition is met.

Do I need to know how to code or write formulas?+

No. Rules are written in everyday language — for example, “Buy when the price closes above the 20-day high and sell when it drops 5% from entry.” Plutux shows the interpreted rules as editable cards so you can confirm it understood you before anything runs.

Is the backtest run on real market data?+

Yes. Backtests use actual historical daily and intraday prices for the stocks and date range you choose, and list every trade the rules would have taken, alongside a buy-and-hold comparison on the same stocks.

Can I test a rule that isn't a number, like one about the news?+

Yes, and this is the part most backtesters cannot do. Alongside price, volume and fundamentals, a rule can ask a question about the news around a stock or about what was said on an earnings call — “exit if the coverage turns negative”, “only buy if management guided up”. Plutux evaluates that question date by date, exactly like a numeric condition, and the same rule keeps running when you paper-trade the system live.

How do you stop the AI from knowing what happened next?+

By cutting the window at the bar being tested. When a rule asks about the news on a given day, the model is shown only articles and transcripts published before that day — never anything after it. Data that exists only as a snapshot of today, such as current ETF holdings or a screener's results, is refused rather than backfilled into history. Hindsight is what turns a mediocre system into a flawless-looking equity curve, so the boundary is enforced in the engine rather than left to each rule.

Does Plutux place real trades?+

No. Plutux paper-trades: it tracks positions your rules would open and close on live prices and keeps score, but it never connects to a brokerage or moves money. You decide what to do with each signal.

Which markets, stocks and timeframes are supported?+

Backtesting and paper trading cover US-listed stocks and ETFs, on bar sizes from one minute to one month. Daily and weekly history goes back a decade or more; intraday depth depends on the bar size. Plutux's research pages separately cover major global markets.

How does the AI turn a sentence into a system?+

Each sentence becomes a chain of cards on a canvas: fetch the data, compute the indicator, check the condition, take the action. You can read and edit every card before a backtest runs. Fully mechanical rules never call the AI at runtime; rules that need pattern recognition or judgement say so, and are labeled by tier.

Is Plutux investment advice?+

No. Plutux is a research and education tool. Market data, backtests and AI-generated analysis are for information only and are not a recommendation to buy or sell any security.

Early access

Build your first system when the doors open.

Plutux is in private testing. Join the waitlist to get an invite, product progress, and the 5-part trading-process checklist.

Plutux is not an investment adviser. Market data and AI-generated analysis are for information and education only, not investment advice. Disclaimer

© Plutux Technology Limited 2026
No-Code Trading System Builder — Backtest, Alerts | Plutux