Categorize, extract, or experiment using AI

We are incredibly excited to announce the launch of three new AI-powered steps, available today to all users!

Parabola has always been powerful for working with structured data. But we know that doesn’t represent all of your tasks – many data workflows are still quite manual and burdensome because they deal with messy, unstructured, or variable data (like PDFs, email bodies, etc.).

That’s where AI comes in. With the combined power of AI and Parabola, you can create structure and automate processes. Use our new GPT-powered steps to transform your data:

  1. Categorize: This step evaluates data and categorizes rows into the groups you define.
  2. Extract: This step extracts specific information from the data you input. By naming the new columns that you want to populate, you tell the step which pieces of info to extract.
  3. Experiment: This step lets you experiment with prompting the AI to revise your input data in any way. It’s the most flexible AI step.

Check out our linked guides for more details. We’re actively improving these steps every day, so please send us your feedback and needs! :brain:


Hey Zach,

How do we get data from a pdf into Parabola?

1 Like

PDF support is coming imminently; stay tuned! :eyes: Someone from our team will reach out to you directly to make sure we stay in close touch about it.

(In the meantime, of course, you can test this out by copying PDF data into any of our other supported formats: CSV, XLS, email body, etc.)

1 Like

Thanks, Graciela! PDF support will be huge for the logistics community. I’ll be happy to share some use cases once I’ve had the opportunity to test it out. I’ve already tested with feature with data from an API request, and it was very impressive.


Absolutely – I imagine your world is flooded with PDFs of all different types!! We’re really excited to get you going with PDF + AI support as soon as possible, and would love to know more about your use cases.

1 Like

We’d be really interested in PDF + AI support for finance/banking/legal applications as well!