Deploy

When you’re confident in your strategy and it has passed backtesting (and ideally cross-validation), it’s time to go live.

1. Verify Your Backtest Results

   Before deploying, make sure your algorithm has:

   • Sharpe > 1.0 — risk-adjusted edge
   • Max drawdown < 20% — manageable risk
   • Consistent equity curve — not dependent on a few lucky trades
   • Cross-validation passing — performs consistently across different time windows

   If you haven’t run a backtest yet, go back to the Backtest Walkthrough.

   Caution

   A strategy that looks great on a single backtest can still be overfit. Run rolling cross-validation from the optimization panel to check robustness before going live. See Backtest Features → Cross-Validation.

2. Switch to Run Mode

   Switch to Run mode using the header mode selector. Open Run Setup, pick your algorithm and configure the run.

   The algorithm executes in real-time using the exact same state machine and logic as the backtester — what you tested is what runs.

3. Pick a Conservative Starting Capital

   For your first live deployment, use a small Starting Capital value. Even a well-backtested strategy can behave differently with real slippage and latency.

   • Start with a fraction of what you intend to deploy long-term (e.g. $100 if you eventually want to run with $1,000)
   • Leverage in the live run can stay the same as your backtest — it sizes positions relative to the Starting Capital
   • You can stop the run and launch a fresh one with a larger Starting Capital at any point

   Note

   Starting with a small notional lets you validate real execution (fills, fees, timing) while keeping risk minimal. Once you see live results matching backtest expectations, stop the run and relaunch with a larger Starting Capital.

4. Monitor Your Live Run

   Run mode gives you dedicated monitoring panels:

   • Run Monitor — real-time view of your active algorithm with position, entry price, equity, trade count, and P&L
   • Run Chart — live chart visualization for the active run
   • Active Positions tab (Info panel) — open positions with real-time P&L
   • Trade History tab (Info panel) — every entry and exit with P&L

   The live engine evaluates conditions on every new candle, checks stop loss and take profit levels, and executes trades automatically.

5. Stop or Adjust

   You can stop your algorithm at any time by setting the run status to STOPPED.

   • Open positions are not automatically closed when you stop — close them manually if needed
   • To iterate: stop the run, switch to Build mode to edit (creates a new version), Test mode to backtest, then back to Run mode to redeploy
   • Use the Agent tools (update_run_status, list_runs) to manage runs programmatically

---

What’s Next?

Want to…	Go to
Connect an AI agent to manage strategies	Agent Walkthrough →
Learn about live deployment in detail	Deploy Features →
Optimize before deploying	Backtest Features → Optimization