Your First Pipeline: Symbol Input → Kalman Filter → Chart Output

Kalmate's Pipeline lets you build visual data workflows by connecting nodes — no code required. In this tutorial, you'll build a live pipeline that fetches AAPL prices, smooths them with a Kalman Filter, and displays a real-time chart.

Step 1: Open the Pipeline Editor

Go to /terminal/pipeline. You'll see a blank canvas with a node palette on the left (Input, Process, Output categories).

Step 2: Add a Symbol Input Node

1. Click Input in the palette to expand it.
2. Drag Symbol Input onto the canvas.
3. Click the node to open its config panel on the right.
4. Enter AAPL as the symbol.
5. Set Interval to 30s (default).
6. Click Save. The node will start fetching live AAPL quotes from Yahoo Finance.

Step 3: Add a Kalman Filter Node

The Kalman Filter smooths noisy price data by estimating the true signal. It's perfect for financial time series where prices jump around.

1. Click Process in the palette.
2. Drag Kalman Filter onto the canvas, to the right of Symbol Input.
3. Click the node. Set Process Noise (Q) to 0.01 and Measurement Noise (R) to 0.1.
4. Q (process noise) = how much the true price can change between steps. Lower = smoother.
5. R (measurement noise) = how noisy the incoming data is. Higher = trust the model more.

Step 4: Connect the Nodes

1. Hover the output handle (right side) of Symbol Input — it glows green.
2. Click and drag to the input handle (left side) of Kalman Filter.
3. A connection line appears. Data now flows: Symbol Input → Kalman Filter.

Step 5: Add a Chart Output Node

1. Click Output in the palette.
2. Drag Chart Output onto the canvas, to the right of Kalman Filter.
3. Connect Kalman Filter's output → Chart Output's input.
4. Click Chart Output, set Chart Type to Candlestick.
5. Set Time Range to 1d.

Step 6: Run the Pipeline

Click the ▶ Run button in the top toolbar. Within seconds:

• Symbol Input fetches live AAPL price (updates every 30s).
• Kalman Filter outputs a smoothed price series.
• Chart Output renders a live candlestick chart.

Come back in a few minutes — the chart updates automatically as new data arrives.

What's Next?

• Add a SMA node after Kalman Filter to compare smoothed vs. simple moving average.
• Swap Chart Output for Candle Chart Output for volume bars.
• Add an Alert Output to get browser notifications when price crosses a threshold.
• Read Advanced Pipeline: Multi-Symbol Mean-Reversion with RSI + SMA (publishes June 28).
• See All 18 Pipeline Nodes Explained (publishes July 5).

Questions? Contact us or open an issue on GitHub.