# Tutorial: Create your first diagram

In this tutorial, you will create a simple diagram representing a common data integration pattern. The pattern includes connecting to an external system, transforming data, building an ontology, and creating a user-facing application.

By the end of this tutorial, you will know how to:

* Add component nodes to your diagram.
* Create connections between nodes.
* Group related nodes together.
* Save your diagram.

## Create a new diagram

1. Navigate to Solution Designer in your Foundry enrollment.
2. Select **New diagram** from the home screen.
3. An empty canvas opens where you can start building your diagram by adding nodes, loading a graph from [Data Lineage](../../data-lineage/overview/index.md), or exploring reference examples.

### Add your first nodes

Follow the instructions in the sections below to build a diagram that shows data flowing from an external system through Foundry and into a Workshop application.

#### Add an external system node

1. Select the **Component** button in the upper left toolbar or press `Cmd/Ctrl + B` to open the component search and selection modal.
2. Select **External system** from the left panel in the component search and selection modal.
3. Choose the relevant **External system** type, such as **Data Store** or **File Store**, and the node appears on your canvas.

#### Add a data sourcing node and connect it to your external system node

Next, add a node to represent how Foundry connects to the external system.

1. Open the component search and selection modal (`Cmd/Ctrl + B`).
2. Search for and add a **Data sourcing** node, such as **Data Connector** or **Data Connection Agent**.
3. Hover over the **External system** node's right connection handle.
4. Click and drag to the **Data sourcing** node's left connection handle.

#### Complete the data flow

Continue building the diagram by adding the following nodes in sequence:

1. **Dataset:** Open the component search and selection modal, search for and select **Dataset**.
2. **Pipeline Builder:** Add a **Pipeline Builder** node to represent data transformation.
3. **Dataset:** Add another **Dataset** node to represent the transformed data.
4. **Object type:** Select the **Resource** button in the top left toolbar, choose **New object node**, and select an object type from your ontology. You can also create a generic object node *not* linked to an existing object type if you prefer.
5. **Workshop:** Add a **Workshop** node to represent the user-facing application.
6. **User group:** Add a **User Group** node to represent a group of users and tasks they need to perform.

![Your first diagram in Solution Designer.](../../../assets/a3be9a37accd8d754dec545492e7c596711711be3db81ef1f053e4099202dd15.webp)

Your diagram now shows a complete data integration and application workflow.

### Group related nodes

To organize your diagram, group the data transformation components together.

#### Create a group

1. Hold `Cmd` (macOS) or `Ctrl` (Windows) and select the **Pipeline Builder** node.
2. Continue holding `Cmd`/`Ctrl` and select the **Dataset** node immediately after **Pipeline Builder**, which represents transformed data.
3. Validate that both nodes display highlighted borders, indicating they have been selected.
4. Select the **Group** button to the left of **Add Text node** in the diagram toolbar.
5. A group container is created around the two selected nodes.

#### Name the group

1. Select the newly created **Group** on your canvas.
2. Choose the default **Group** label to provide the group with a descriptive name, such as `Data Transformation`.
3. Select anywhere on your canvas outside the group's borders and **Save** your changes.

Your diagram now has a logical grouping that makes it easier to understand your workflow's architecture.

### Add descriptions and context

#### Add node descriptions

You can add notes to individual nodes to document their purpose:

1. Select the **Pipeline Builder** node.
2. Choose **Enter description...** and add a description, such as: `Cleanses and transforms raw data into business-ready datasets`.
3. Repeat this process for other key nodes in your diagram.

#### Add a text node for comments

To add general comments to your diagram:

1. Select the **Add Text node** button in the upper left toolbar, and a text node appears on your canvas.
2. Select **Double click to edit** to edit the text.
3. Enter a description of your diagram, such as: `This diagram represents our core data integration pattern for external vendor data`.
4. Select and drag the text node to position it near the top of your diagram.

### Save your diagram

Now that your diagram is complete, save it to the Foundry filesystem:

1. Select **Save** in the Solution Designer header.
2. A **Save** modal renders if this is your first save, where you can enter a **File name**, such as `First Diagram`, and choose a save **Location**.
3. Select **Save** to complete the save process.

![Completed diagram in Solution Designer.](../../../assets/28097761da7128dbd1fed60aeeeb2dfc0de6c3a14cc7e82ba81743339b45a932.webp)

Your diagram is now saved and will auto-save as you make future changes. Select **Draft saved** from the file header to view a modal with additional information about your auto-saved draft.

## Next steps

Now that you have created your first Solution Designer diagram, continue exploring its features and capabilities by:

* **Exploring reference patterns:** Select **Reference Diagrams** in the navigation bar on the left side of your screen to browse pre-built industry and technical patterns.
* **Linking resources:** Select nodes in your diagram and use the **Link resource** button in the node toolbar to connect them to actual Foundry datasets, pipelines, or applications.
* **Using AIP Architect:** Try [AIP Architect](../aip-architect/index.md) to generate workflow plans based on your requirements.
* **Reviewing your diagram:** Use [AIP Critic](../aip-critic/index.md) to get AI-powered feedback on your architecture.
* **Customizing your diagram's view:** Open [**Graph Settings**](../navigation/index.md#configure-your-diagrams-settings) to experiment with different display options, such as icons-only mode or edge animation.

For more details on diagram features and capabilities, see [Diagrams](../diagrams/index.md).

## Tips

* **Start conceptually:** Begin with high-level components and refine later. You do not need to specify exact datasets or pipelines in your first draft.
* **Use groups liberally:** Group nodes by function (for example, "Data Ingestion", "Processing Layer", "Application Layer") to make complex diagrams easier to navigate.
* **Leverage patterns:** Browse the reference architecture library for proven patterns that match your use case. You can load a pattern and customize it rather than starting from scratch.
* **Iterate with feedback:** Use AIP Critic to validate your design, then refine based on suggestions.
* **Think about layers:** Organize your diagram in logical layers from left to right (data sources → ingestion → processing → ontology → applications → users) for easy comprehension.