Collaboration in Dataiku

Dataiku = Your Place to Work on Data With Your Team

Dataiku was built with collaboration front and center. You may be starting a project on your own, but it’s always easier to start thinking about collaboration from the start than having to document, rename, or rebuild things later.

If you aren’t planning to share your work with anyone, most of these features will help you understand your project better when you return to it.

The Flow

Your Whole Project in One Glance

Looking at a project Flow, you can trace back all the steps in a project: the datasets and the recipes. Use naming to make your recipes more straightforward to understand and jump in. Click on an object to access the Details tab on the right with additional information.

You can copy sections of your Flow, including into another dataset, to reuse work across projects.

Tip: Use the Explain Flow recipe to leverage Generative AI to explain your Project Flow.

Flow Zones

Group Recipes to Make Your Flow Easy to Read

Create Flow Zones to group several elements in your Project Flow, recipes, and datasets. In the Actions tab, under Flow Zones, select Move (or right-click on the selection and select Move to a flow zone), then name your Flow Zone.

Naming

Save Time With Naming and Descriptions

Good dataset naming helps you and your colleagues quickly understand what a Flow achieves. Ideally, dataset names should be readable, self-explanatory, and short, with consistent naming rules. You can also add descriptions to most objects in your project, which can be read from the flow.

Tip: You can rename a dataset in the Flow by right-clicking on it to open the context menu or finding the same function in the right panel.

Tags

Tag Objects to Navigate More Easily

Tags in Dataiku are available on many objects. You can tag projects, or objects in a project such as datasets, recipes, models, or even wiki articles. You can create tag categories.

Dashboards

Share Charts, Insights, and More

Dashboards are the key to communicating on a project: sharing charts, datasets, information, and even buttons to run scenarios and build datasets. To create one, click the dashboard tab in your project’s top left navigation. You can share a dashboard with users who don’t have access to your project.

A dashboard is made up of one or more slides. Each slide is made up of tiles, which you can drag and drop on a grid as needed. Each tile on a dashboard slide holds an insight from the project.

There are many types of insights, such as charts, model reports, Jupyter notebooks, web apps, metrics, macros, etc.

Workspaces

Package Information to Share With Your Team

Create a Workspace to communicate information easily. These work as a private space and single point of access where people can easily find, review, and discuss all the Dataiku assets they need related to a business initiative, for example, even if they’re sourced from different projects.

When you share be mindful of access rights.

Project Wikis

Create Documentation with Wikis

You can build a whole wiki on a project. These can be very sophisticated wikis with multiple articles, a table of contents, links to objects, etc. To get started, navigate to the Wiki image in your project navigation and click to create your first article.

Start with a sample Wiki that includes guidelines for documentation in Dataiku.

Discussions

Start Conversations on Dataiku Objects

On most objects, you can click the chat icon to launch a discussion with one or several colleagues within Dataiku. You can have several threads on an object, and mark them as resolved. Discussions you’re involved in will appear in your notifications center.

Version Control

Roll Back to a Previous Version of Your Project

Sometimes, somebody messes with your work, and you need to revert. Sometimes, that somebody is you. The good news is you can quickly revert to a previous version of your project by going into Version Control in your top project navigation.

We’re happy you’re here

Join a Community of Dataiku users

Have a question? Want to explore Dataiku tips? Share an idea you have to improve Dataiku? Explore great stories from our users?