QuantDinger Tutorial: Backtest and Deploy Your First AI Trading Strategy

In this tutorial, you'll go from zero to a running QuantDinger instance, then backtest a simple moving-average crossover strategy and connect a paper-trading account. Everything runs in Docker, so you only need a machine with Docker and about 4 GB of RAM.

Step 1: Start the Backend

Create a project directory and a docker-compose.yml that pulls the official image ghcr.io/openbyteinc/quantdinger-backend:latest. Map port 8080 and mount a volume for persistent data. Run docker compose up -d and wait for the health check — the web UI will be available at http://localhost:8080.

Step 2: Configure Market Data

From the dashboard, add a data source. QuantDinger supports major exchanges out of the box:

Step 3: Run Your First Backtest

Create a strategy — either from the built-in templates or the multi-agent research panel. Pick a symbol and date range, then hit Backtest. QuantDinger returns equity curves, drawdown, win rate, and Sharpe ratio. Iterate on parameters until the results match your risk profile.

Step 4: Go Live With Paper Trading

Connect an Alpaca or exchange account in Settings → Exchanges. Start with paper trading: the platform executes your strategy against simulated balances so you can verify behavior without financial risk. Once you're confident, flip the account to live mode.

Step 5: Monitor and Extend

QuantDinger sends Telegram notifications for trade events and lets you hook in the MCP server so AI assistants can query positions or trigger research runs. The platform is Python-based and Apache-2.0 licensed, so extending it with custom indicators or agents is straightforward.

That's it — you now have a self-hosted AI trading stack running 24/7. Start with paper trading, validate your edge, and scale up gradually.