Skip to Content

Projects

Projects are the top-level organizational unit in Rhesis: each project groups the endpoints, tests, test sets, and results for one LLM application or testing initiative.

Prefer code? Create and manage projects with the Python SDK (entities).

Understanding Projects

A project can hold multiple endpoints, so you can test the same LLM application across environments (development, staging, production) or compare different implementations and API configurations.

Parameter Schema

A project also owns a Parameter Schema, which defines the typed configuration slots your application supports (like model selection, temperature, and system prompts). Within the project, users create and manage Experiments that bundle values for these slots.

Working in a Project

Most Rhesis views are scoped to the active project. The frontend sends the active project as X-Project-Id, and backend queries use that project scope for endpoints, tests, test sets, test runs, traces, and related records.

The sidebar shows the active project name with the organization name underneath. Open the organization menu and choose Switch project to select a different project from a searchable drawer.

Project switcher behaviorDetails
Single projectAutomatically selected
Multiple projectsChoose from project memberships in the switcher drawer
PersistenceStored in a browser cookie and saved as your default_project setting
SwitchingReloads the page so every view refetches with the new project scope
No membershipsShows No project access with create-project and refresh actions

If a grid appears empty after switching workstreams, confirm the active project in the sidebar before changing filters.

Creating a Project

Open the organization menu in the sidebar, choose Projects, and click the + (Create project) button.

Once your project is created, you’ll typically:

  1. Add endpoints that connect to your LLM application’s API
  2. Create or generate tests to validate your AI behavior
  3. Organize tests into test sets for execution
  4. Run tests against any of your project’s endpoints and analyze results

Managing Projects

Projects Overview Page

Click a project card to open the project details page, where you can edit settings or delete the project.

Search, filters, and pagination

The Projects page supports discovery controls for larger organizations:

ControlBehavior
SearchFilters by project name and description
Status filterToggle between All, Active, and Inactive
PaginationNavigate results with configurable page sizes (25, 50, 100)

These controls apply together, so you can search within a status-filtered subset of projects.

Project Status

Projects can be marked active or inactive from the project edit drawer. The status drives the badge on the project card and the status filter on the Projects page, so you can keep finished initiatives out of view without deleting them.

Projects Detail Page

Deleting Projects

Important: Deleting a project is permanent. Consider marking it inactive instead if you want to preserve historical data.

To delete a project, click the delete icon on its card and confirm in the dialog. This permanently removes the project and all associated endpoints, tests, and test sets.


Next Steps - Add Endpoints to connect your AI application - Create Tests to validate your AI behavior - Organize tests into Test Sets for execution