SignalAI
MarketplaceYour signalsStrategiesBacktestsPaper trading SoonMCPDocs

Your first backtest

From zero to a backtest you can see, in about five minutes. (Engine details are handled for you — you just write Python.)

1. Install + scaffold

curl -fsSL https://signal-ai-mu.vercel.app/install.sh | sh   # the signalai CLI
pip install signalai-quant                                   # the Python library
signalai login                                               # opens your browser to sign in
signalai init my_strategy.py

2. Write a strategy

A tiny mean-reversion idea: buy when a bar closes red, sell when it closes green. You subclass Strategy, subscribe to bars, and place orders — you never touch an engine.

from signalai_quant import Strategy, Bar, Side, OrderType

class MyStrategy(Strategy):
    def init(self):
        self.subscribe(Bar, self.on_bar)

    def on_bar(self, bar):
        held = self.portfolio[bar.symbol].qty
        if bar.close < bar.open and held == 0:
            self.order(bar.symbol, Side.BUY, 10, OrderType.MARKET)
        elif bar.close > bar.open and held > 0:
            self.order(bar.symbol, Side.SELL, held, OrderType.MARKET)

3. Submit it

Backtest it in the cloud over years of real data:

signalai submit my_strategy.py --symbols AAPL --from 2015-01-01 --to 2024-12-31
signalai status <id>

When it finishes, view the equity curve, trades, and stats at your backtests:

  • equity curve — account value over time
  • trades — every fill (timestamp, side, qty, price)
  • stats — return, max drawdown, Sharpe

What to try next