A project is a workspace for your AI applications. Projects maintain logical separation for different AI use cases, typically within the same team. Projects house different experiments and apps.

Scenarios where you should create different projects

• You are testing a new LLM application in development, and want to maintain separation from your current production application.
• You have a new LLM use case, which will serve a different use case from other AI application projects.
• You are creating a demo project for internal or external use.
• Your team mantains several AI system architectures, and you want to compare performance across them over time.
• Different team members are contributing to different AI projects internally.

Create a project

To run an experiment, first create a project. Multiple experiments can be run in the same project. You can create a new project in the UI in Experiments -> Projects. You can obtain the corresponding Project ID in the URL by clicking on the project. This is in the format https://app.patronus.ai/projects/project-id.

You can also create projects with the https://api.patronus.ai/v1/projects endpoint. See the API reference guide for details.