Announcements
REMINDER: Sign up for the Foundry Newsletter to receive a summary of new products, features, and improvements across the platform directly to your inbox. For more information on how to subscribe, see the Foundry Newsletter and Product Feedback channels announcement.
Share your thoughts about these announcements in our Developer Community Forum ↗.
Introducing custom website hosting in Developer Console for Developer Tier enrollments
Date published: 2025-01-16
We are excited to expand the availability of website hosting to Developer Tier enrollments. This powerful addition enables you to seamlessly build and deploy custom frontend applications, leveraging Foundry as your backend with the help of the Ontology SDK (OSDK).
To get started, navigate to Developer Console and create a client-facing application. See the documentation for an in-depth walkthrough:
Provision custom subdomains
Use Developer Console to provision a custom subdomain. Foundry will automatically provision the required infrastructure including DNS records and TLS certificates. Your application will be served from one of the following domains depending on your enrollment type:
- Enterprise tier enrollments:
{APPLICATION-NAME}.[YOUR-ENROLLMENT].palantirfoundry.com - Developer tier enrollments:
{APPLICATION-NAME}-[HASH].apps.[ZONE].palantirfoundry.com
The website hosting menu in Developer Console.
With this feature, Developer Tier enrollments can now take advantage of Foundry's development tooling and website management tools.
Kick-start development
Go from an empty slate to a deployed website in 10 minutes. Start with a custom repository template, develop inside a fully set up VS Code workspace, and manage your changes in Code Repositories.
The development environment in Foundry, with the VS Code IDE on the left and live preview of the application in development on the right.
Continuous deployment
Leverage robust version control tools provided by Code Repositories to review and safely deploy your website changes.
The version control page in Code Repositories.
Website management tools
Configure your website's content security policy, manage roll-back, and monitor usage metrics.
Developer Console provides tools to monitor your application's Foundry usage.
Introducing manual entry transform tables in Quiver [GA]
Date published: 2025-01-16
Generally available across enrollments the week of January 13, you can add manual entry transform tables as cards to your Quiver canvas, enabling you to create a transform table from scratch that contains up to 5,000 rows of data. Manual entry transform tables have an intuitive spreadsheet-like user interface and support five data types: string, number, Boolean, time, and time series.
As with Quiver's other transform tables, you can apply any of the transform operations available in the transform table search window to manual entry transform tables.
When should I use manual entry transform tables?
As an example use case, you can create a manual entry transform table to dynamically parameterize an analysis in conjunction with row and column selectors. The values in the table's selected rows can be used as dynamic parameters downstream, such as the figures for metric_b in the Time Series Chart on the Quiver canvas in the image below.
Use manual entry transform tables to parameterize an analysis in Quiver.
You can also use manual entry transform tables to ingest small sets of data from external sources to supplement an analysis and integrate with the Ontology. Additionally, manual entry transform tables enable the full range of Quiver's time series analysis operations for time series datasets containing up to 5,000 rows without the need to establish a time series sync.
To learn more about manual entry transform tables and the workflows they support, see Quiver's transform table documentation.
Benefit from granular control over Project resource queue assignments with the Resource Management application
Date published: 2025-01-16
Available today on all enrollments, you now have new options in the Resource Management application to give you more granular control over a Project's resource queue assignments. In the past, GPU-enabled Projects that were assigned to vGPU resource queues used their enrollment's default resource queue for vCPU workloads. Now, you can configure these types of Projects to use a specified vCPU resource queue. This means that Projects can be assigned to up to two different resource queues, a vCPU queue, and optionally, a vGPU resource queue, to give you more flexibility over your resources.
Additionally, you can now support critical workflows that require dedicated compute resources by making one existing branch per Project the "priority" branch for that Project. Workloads on a set priority branch benefit from the ability to use assigned resource queues, as opposed to other workloads which continue to use the resource queues assigned to the Project. Like Projects, each priority branch is assigned to a vCPU resource queue, and they can also optionally be assigned to a vGPU resource queue.
Enable priority branch and manage which vCPU and vGPU resource queues the priority branch uses.
You can now also view a Project's resource queue assignments in the platform filesystem sidebar. For example, the following screenshot demonstrates a Project with a priority branch namedproduction, with distinct vCPU and vGPU resource queue assignments between the priority branch and the Project itself.
View a Project's resource queue information at a glance from the platform filesystem sidebar.
For more information, visit the priority branch documentation.
Use Python transforms in VS Code workspaces [Beta]
Date published: 2025-01-14
We are excited to announce that you can now develop Python transforms in VS Code workspaces, allowing you to use the Visual Studio Code IDE to seamlessly develop your Python transforms with Palantir workflows. This beta feature is available to all users who have access to VS Code workspaces (also in beta).
The VS Code workspace interface, featuring improved developer tools.
Why should I develop my Python transforms in VS Code?
Python transforms with VS Code workspaces unlock significant improvements to tooling and capabilities that meaningfully improve developer workflows. You can expect the following updates to your developer experience:
- Fully customizable IDE: Customize the VS Code workspace IDE to your preference, with themes, keybindings, interface layouts, and more.
- Palantir extension for Visual Studio Code: This extension bridges VS Code and Foundry, allowing you to perform and use Python transform operations (such as transforms preview, debugging, and the library panel) natively in VS Code.
- Full dataset (sample-less) preview: Rather than running a full build, you can run previews on the full dataset to get results.
- Performant language server: Benefit from fast autocomplete suggestions and linting.
- Transforms type hints: Use a transforms-aware linter that suggests correct transforms parameter types.
- Optimized environment setup: Use Palantir’s latest high-performance environment management tool to set up your Python environment quickly and efficiently.
- Terminal access: Use a powerful and efficient tool to run shell commands. For example, you can run a Python interactive shell or git CLI.
- Improved source control: Leverage VS Code’s built-in source control interface or the git CLI in the terminal for complex source control operations.
Get started
To start using Python transforms in VS Code workspaces, open your transforms repository in the Code Repositories application. From here, select Open in VS Code from the top right corner of the screen, which will take you to a VS Code workspace:
Open your Python transforms code repository in a VS Code workspace by selecting Open in VS Code.
VS Code workspaces are in a beta state and available by default in all Organizations where Code Workspaces is enabled. If you do not see the option to open your code repository in a VS Code workspace, contact Palantir Support to learn how to enable access.
Limitations
Python transforms in VS Code workspaces is a new feature; some transforms preview components are not yet supported. We are actively working to add support for all missing features. For more information, review our documentation.