> ## Documentation Index
> Fetch the complete documentation index at: https://docs.cord.nyc/llms.txt
> Use this file to discover all available pages before exploring further.

> ## Agent Instructions
> You are helping someone use Cord, the capital and resource allocation platform.
> Use title-case Cord in prose. Prefer workspace, project, ontology, lineage, scenario, and allocation.
> Say source-backed or sourced for answers with lineage. Do not say hallucination-free.
> Product URLs are https://app.cord.nyc, https://cord.nyc, and https://trust.cord.nyc. Support is support@cord.nyc.
> Give concrete product steps. Do not invent UI labels, CLI commands, or unreleased agent runtime details.
> Use Understanding Cord for concepts and Guides for task steps.

# Prepare data with a pipeline

> Clean, combine, and reshape your source data into a reusable output.

<Badge className="cord-ai-badge" color="gray" shape="rounded">
  <span className="cord-ai-badge-copy">
    <span className="cord-ai-badge-question"><Icon icon="sparkles" size={16} />Don't want to do this all manually?</span>
    <span className="cord-ai-badge-description">Cord Computer can do this for you.</span>
  </span>
</Badge>

A pipeline connects sources to a series of steps, called nodes. You can preview the data after each step and create separate outputs from different branches.

1. Open [**Pipelines**](https://app.cord.nyc/pipeline-studio) with your project selected.
2. Select **New pipeline**, enter a name, and select **Create pipeline**.
3. Choose a data feed or project file as your source.
4. Use the **+** on a node to add a step, such as **Clean**, **Filter**, or **Group By**. Select the node to configure it.
5. To combine sources, use **Add source**, then add a **Join** step and choose the inputs and matching columns.
6. Select a step and choose **Run to here** to check its rows in the preview.
7. Add a **Pipeline Output** node at the end of a branch. Select it and choose **Create artifact** to save an output version.

Your output is ready to preview, export, or [write to your ontology](/write-pipeline-data-to-ontology).

<Frame>
  <img className="block dark:hidden" src="https://mintcdn.com/cord/Ov8oNa6INQN9kYdr/images/guides/prepare-data-pipeline-light.png?fit=max&auto=format&n=Ov8oNa6INQN9kYdr&q=85&s=2f6ff1d7bcf25d9cf7a1c4f679d15b0f" alt="A 12-node demand planning pipeline with three sources, two outputs, and a buyer summary preview" width="1600" height="1000" data-path="images/guides/prepare-data-pipeline-light.png" />

  <img className="hidden dark:block" src="https://mintcdn.com/cord/Ov8oNa6INQN9kYdr/images/guides/prepare-data-pipeline-dark.png?fit=max&auto=format&n=Ov8oNa6INQN9kYdr&q=85&s=bdcc51424b3c74cac119b6d5399ea14e" alt="A 12-node demand planning pipeline with three sources, two outputs, and a buyer summary preview" width="1600" height="1000" data-path="images/guides/prepare-data-pipeline-dark.png" />
</Frame>

## Node types

| Node              | Use it to                                                                         |
| ----------------- | --------------------------------------------------------------------------------- |
| Source            | Load data from a feed or project file.                                            |
| Filter            | Keep or remove rows that match your conditions.                                   |
| Clean             | Fix values and column types, fill blanks, or remove duplicates.                   |
| Calculated Column | Add columns using formulas or existing values.                                    |
| Join              | Combine two tables using matching columns.                                        |
| Pivot             | Turn categories into columns, columns into rows, or transpose a table.            |
| Group By          | Summarize rows with totals, averages, counts, or other grouped values.            |
| Split             | Turn multiple values in one cell into separate rows.                              |
| Case              | Set a column's value using the first matching rule.                               |
| Window            | Compare rows using previous or next values, percent changes, or rolling averages. |
| Python            | Run custom Python code on incoming rows or source files.                          |
| AI Assist         | Use a prompt to classify or enrich rows in a new column.                          |
| Visualization     | Preview data as a chart or table without changing it.                             |
| Pipeline Output   | Save an output version for preview, export, or downstream use.                    |

Select the info icon (**Node guide**) in the pipeline toolbar for options and examples.