Scout AI Trading: From Plain-English Idea to Backtest

Scout is Polarline Markets' AI trading research assistant. You describe a trading idea in plain English; Scout writes a complete Python strategy, backtests it against historical data, and helps you refine it. The strategy code stays inspectable, so you can read every rule before you trust a result.

Two ways to use Scout for AI trading research

• The Scout CLI. The open-source terminal app (see Scout CLI). Strategies and backtests run locally with your own model key.
• Scout Cloud by Polarline. The browser workspace for investment research, strategy development, and backtest evidence (see Scout Cloud).

How a strategy is built and tested

• Generated strategies follow the Shape-C contract, with rules that prevent lookahead bias.
• The engine_v2 backtester reports 60+ metrics, with walk-forward tests, parameter sweeps, and Monte Carlo.
• Bring your own model: Claude, OpenAI, Gemini, or a local model (see BYOK).
• Brokers are optional: Alpaca, Binance, and Interactive Brokers (see brokerages).

Get started

curl -fsSL https://www.polarline.io/cli/install | bash
finny

The Free plan is $0 for the open-source CLI. Plus is $20/month and Pro is $50/month. See Polarline pricing.

Polarline Markets team