Back to blog

Dev Diary #7 - Exploring Versions


This week we have added the ability to explore the data in a dataset directly in the browser, and we have also been working on a queue system to enable some new features.

Samples Explorer

Today is the last day you will need to download a dataset and convert it to a human readable format because we have added the ability to view the dataset directly in the browser. While it is still fairly basic, you can now see the all of the samples (the input output pairs) in a dataset, and we will be adding more features to this view in the future.


To make the platform more usable, we have been working behind the scenes to enable a couple of often-requested features.

  • Create and Train
  • Multiple Edits
  • Clustering

The common thread is that they require us to setup a queue system to handle them. While we are still working on the UI changes for these features, the backend is now ready to handle them.


  • View individual samples in a dataset version
  • Fixed a bunch of bugs related to schema editing
  • Big styling pass across the entire app for better consistency
  • App is now more responsive and looks much better on non-standard dimensions