# Agents in the AM

> Agents in the AM - A walkthrough for creating an agentic workflow, configuring metrics, and chatting
> with it.

This Markdown file sits beside the HTML page at the same path (with a `.md` suffix). It summarizes the topic and lists links for tools and LLM context.

Companion generated at `2026-05-06T18:17:09.943944+00:00` (UTC).

## Primary page

- [Agents in the AM](https://docs.datarobot.com/en/docs/get-started/how-to/agents-am.html): Full documentation for this topic (HTML).

## Sections on this page

- [1. Log in and navigate to Workbench](https://docs.datarobot.com/en/docs/get-started/how-to/agents-am.html#1-log-in-and-navigate-to-workbench): In-page section heading.
- [2. Create a codespace](https://docs.datarobot.com/en/docs/get-started/how-to/agents-am.html#2-create-a-codespace): In-page section heading.
- [3. Enable port forwarding](https://docs.datarobot.com/en/docs/get-started/how-to/agents-am.html#3-enable-port-forwarding): In-page section heading.
- [4. Rundr start](https://docs.datarobot.com/en/docs/get-started/how-to/agents-am.html#4-run-dr-start): In-page section heading.
- [5. Run the application](https://docs.datarobot.com/en/docs/get-started/how-to/agents-am.html#5-run-the-application): In-page section heading.
- [6. Deploy the agent](https://docs.datarobot.com/en/docs/get-started/how-to/agents-am.html#6-deploy-the-agent): In-page section heading.
- [That's it!](https://docs.datarobot.com/en/docs/get-started/how-to/agents-am.html#thats-it): In-page section heading.
- [Try this next](https://docs.datarobot.com/en/docs/get-started/how-to/agents-am.html#try-this-next): In-page section heading.

## Related documentation

- [Get started](https://docs.datarobot.com/en/docs/get-started/index.html): Linked from this page.
- [How-tos](https://docs.datarobot.com/en/docs/get-started/how-to/index.html): Linked from this page.
- [agentic workflows](https://docs.datarobot.com/en/docs/agentic-ai/index.html): Linked from this page.
- [Talk to my Data Agent](https://docs.datarobot.com/en/docs/get-started/how-to/talk-data-walk.html): Linked from this page.
- [GenAI basic](https://docs.datarobot.com/en/docs/get-started/how-to/genai-walk-basic.html): Linked from this page.

## Documentation content

This how-to illustrates the ease of creating [agentic workflows](https://docs.datarobot.com/en/docs/agentic-ai/index.html) in DataRobot using the [Agentic Starter application template](https://github.com/datarobot-community/datarobot-agent-application). After you complete it, try some additional how-tos, listed [at the end](https://docs.datarobot.com/en/docs/get-started/how-to/agents-am.html#try-this-next).

In this walkthrough, you will:

1. Open a codespace and create a local environment.
2. Create and test an agent.
3. Deploy the agent to DataRobot.

## 1. Log in and navigate to Workbench

1. Log in to DataRobot; you will land on the home page.
2. Navigate toWorkbenchand create a new Use Case.
3. Name it “Agents in the AM”.

## 2. Create a codespace

1. From the Use Case getting started page, navigate to codespaces by clickingNotebooks & codespaces:
2. ClickCreate codespaceto create a new codespace.

## 3. Enable port forwarding

Since you are developing within a DataRobot codespace, you need to expose the development ports. This is configured in the Exposed Ports section of your Session Environment tile.

1. Click theSession Environmenttile.
2. In theExposed Portssection, click+ Add port.
3. Enter5173and click the checkmark icon to save.

## 4. Run dr start

1. Click Start a session to start a new session.
2. Once the session has started, click theTerminaltile.
3. In the terminal, typedr startand pressEnter. drstart
4. SelectAgentic Starterand pressEnter. Understanding thedr startcommandIf you are not already inside a clone of the template, thedr startwizard walks through template selection and clone process, then configures the application and creates a.envfile. For the full step-by-step wizard, seePrepare your local development environmentin the template README.
5. For the next several steps, pressEnterwhen prompted to accept the default values configured by the template. This includes the destination directory name, session secret, OAuth provider, Pulumi passphrase, Use Case ID, LLM configuration, and MCP configuration.
6. PressEnterto finish the wizard.
7. Specify whether you wish to use the NeMo Agent Toolkit by selectingyorN. If you do not, specify which agent framework you wish to use by selecting from the list of available frameworks and pressEnter. Using the NeMo Agent ToolkitThe NeMo Agent Toolkit is a low-code YAML-based framework for building agentic workflows. It provides a structured foundation that supports multi-agent workflows, state management, and complex agent orchestration patterns.
8. When prompted to choose a stack, pressEnterto create a new stack. Pleasechooseastack,orcreateanewone:[Usearrowstomove,typetofilter]><createanewstack>
9. Enter a stack name and pressEnter. AITAM_<LASTNAME>

When your terminal returns to the prompt, you are ready to run the application.

## 5. Run the application

1. Navigate to the template directory: cddatarobot-agent-application
2. Start the application frontend, backend, agent, and MCP server: drrundev
3. When the services are ready, open the Agentic Starter interface by clicking the link alongside the port in theExposed Portssection.
4. In the new tab, send a test message to the agent.

> [!TIP] Tip
> Initial output generation can take a few minutes, so while it is "thinking," learn about [syftr](https://www.datarobot.com/blog/pareto-optimized-ai-workflows-syftr/), DataRobot's open-source framework for searching agentic workflow configurations to determine the optimal structure, components, and parameters for your data and use case.

1. When you've finished testing, return to the terminal and press Ctrl+C to stop the application.

You have successfully created an agent! If you like, you can now proceed to deploy the agent to DataRobot.

## 6. Deploy the agent

1. Ensure you are logged in to Pulumi by running pulumi login --local or log into your Pulumi Cloud account.
2. Deploy to DataRobot by running dr run deploy .

> [!NOTE] Updating during deploy
> If you are prompted to perform an update during deployment, select `yes` to accept and perform the update.

Deployment takes several minutes. When it completes, the command prints outputs such as deployment IDs, chat endpoints, and the Agent Playground URL. See [Deploy your agent](https://github.com/datarobot-community/datarobot-agent-application#deploy-your-agent) in the template README for details.

## That's it!

Congratulations—you now have a working agent. Feel free to leave the platform; the codespace will terminate by itself. Find your agent at any time in the Workbench Playgrounds tile.

## Try this next

Continue with other agentic walkthroughs on the [How-tos](https://docs.datarobot.com/en/docs/get-started/how-to/index.html) page, such as [Talk to my Data Agent](https://docs.datarobot.com/en/docs/get-started/how-to/talk-data-walk.html) or [GenAI basic](https://docs.datarobot.com/en/docs/get-started/how-to/genai-walk-basic.html).