Project Definition

Starting June, users will be upgraded to the new version of Project Definition. You can find the documentation for that version here.

Definition is the starting step where the labelling task is created. You'll upload your data, create a taxonomy, and write instructions.

Selecting Project Type

Create a new project by hitting the "+ Create new project" in the dropdown.

You'll be brought to this setup flow, and you'll need to complete these four steps to finish defining the initial state of your project. If it's your first time on Rapid, it is worth browsing the project templates to see examples of data, taxonomy, and instructions.


First time? Start with a template. Otherwise, fill out the information manually.

Starting with a template (Optional)

Once you click Start with a template you'll be brought to a modal to explore a variety of templates across project types.


Browse templates to see examples of the task, instructions, and taxonomy

You'll need to select a Project Type that fits your needs. You can see a preview of an example project and a short description of it. Once a project type is set, you can't change it due to the nature of the project type determining what features are available. You can, however, create a new project with a different type.

For complete details on all available task types, please refer to our API Documentation.


Browse project types and see examples of them. Pictured here is '2D Semantic Segmentation'.

Upload Your Data

The next step is to upload some data relevant to your project. Initially you just need some data to get started - you don’t need to upload all your project data now. You can always upload more later on.


Click Upload data to open up the upload window

You can upload the data with the file type supported for your Project Type.

For example: You would need to upload images for Image Annotation projects and PDFs for Document Transcription projects


Do a local upload or select cloud type to import from.

Once you have uploaded data, it will show in this grid


Query by your metadata to browse your data

Create Project Taxonomy

The next step is to define your project’s taxonomy. This is a JSON string that represents the structure of the data you’re looking to annotate. Access this by clicking Taxonomy from the setup flow.


Click Taxonomy to create the structure of your data

For more information about how to write your project taxonomy for your specific Task Type, please refer to our Documentation.

Write the JSON keeping in mind that the taxonomy you create will structure the UI that the taskers will be using the label your data. Rapid lets you work on the JSON and preview how it will look in the labelling interface side by side. Check that it makes sense before you move on.


Taxonomy editor allows you to write the JSON alongside an example item from your uploaded data.

Write Instructions for Labelers

The final step in project setup is writing task instructions.


Click Task instructions to open the instructions editor

Writing instructions is a crucial part of setting up your project. Thorough instructions will result in quality data because everything will be laid out clearly for the taskers to follow.


The instructions editor will be pre-populated with a template based on the taxonomy you've created as well as any templating you've selected.

Instructions should be thorough and be clear enough for taskers to follow

  • The overall goal of the task
  • Key concepts the tasker should understand
  • What the entities being labeled are
  • Sufficient examples that are representative of the overall dataset
  • Any nuances or special cases they might encounter

Note that instruction writing is an iterative process. We’ve built in a feedback system for our taskers to give you feedback on your instructions.

Updated 4 months ago