Coming from v3 / Cockpit

Are you already familiar with AX and the Cockpit? Then this guide will help you understand what changed in Studio and how Pipelines simplify your data processing.

Studio vs Cockpit

Throughout the reference documentation you will find information boxes like this that tell you about differences between Cockpit and Studio.

What's new in Studio?

We have introduced a new concept called Pipelines, which replace projects as an orginization structure. With pipelines we drastically reduce the amount of up-front data processing, management and the uploading of lots of duplicate data. You no longer need to upload your data multiple times into different collections or projects just to get texts in different languages or for different use-cases. Collections as you know them are gone completely, and you are now able to upload your data only once and generate many different texts from it.

Learn more about Pipelines and how they work in the Pipelines documentation.

The completely new Preprocessor allows you to transform your data into a format that is ready for text generation. This is a powerful tool that can be used to clean up your data or transform it into a format that is easier to work with. You no longer need to transform your data beforehand, but can do this right inside AX Studio now.

With our new Fanout feature you can now define how each data object is generated. This allows you to generate different texts from the same data, for example in different languages or for different output channels.

Full Ruleset Compatibility

How you write rulesets has not changed in Studio. The Composer UI works exactly the same as before, and you can import existing rulesets from Cockpit into Studio pretty easily and without any changes. The only change is that we have renamed rulesets to Blueprints and you no longer have exactly one ruleset per project, but can have multiple blueprints per pipeline.

Full List of Changes and how Cockpit concepts map to Studio

Team -> Organization

"Team" has been renamed to "organization", but fulfills the same role of organizing users and giving them access to pipelines. Your subscription and billing are bound to your organization.

Project -> Pipeline

Pipelines replace projects as an organizational unit, but are much more powerful.

Ruleset -> Blueprint

Rulesets have been renamed to blueprints, but are still the same concept, with the same UI and functionality.

Collection -> Data Pool

Collections are no longer needed to control in which language a text is generated. If you have multiple different data types that you don't want to store in the same place, you can use data pools to separate them. Don't use data pools as a replacement for collections though! If you had a lot of collections in Cockpit, you will not need that many data pools in Studio. For example, if you purely have product data and no secondary data like a list of brands, you will only need one data pool.

Document -> Upload + Object + Result

In Cockpit, a document held both your uploaded data and the generated text. In Studio, we have split this into three parts:

• Uploads: Your raw data, uploaded to the pipeline. We keep a historic record of your uploads.
• Objects: The data you uploaded and preprocessed, in the form of objects. Each object represents a single unit of data that can be used for text generation.
• Results: The generated text, stored in the results storage. Since each object can have multiple generation results, we store them separately and also keep a historic record of text generations.