Fresh off OpenAI's first developer conference, DevDay, we asked Harsh Pareek, Senior Machine Learning Engineer, to talk about what OpenAI announced and how Prodigal is using it right away.

New APIs from OpenAI

At DevDay, OpenAI announced lots of new features. They announced new versions of old models.

And the two most exciting things they announced are Vision API and Assistants API.

With Vision API, you can give any image to OpenAI and you can ask questions based on the image.

How Prodigal is using OpenAI's Vision API right now

So the way we're using it is we're using it to rate email templates and rate landing pages.

We have an internal service where you can forward us your email template, you can forward us your landing page template, and we'll rate it for various factors, like how mobile-optimized it is, you know, what's, like the urgency of your message? What's your call to action?

How you can benefit

So lenders could use this service that we offer for free to understand how consumers will perceive some of these things and engage with the consumers better.

And it's not just that.

OCR software has always been really difficult to work with. It really struggles to work with things like tables.

So you can give us your email templates and such, but you can also actually give us any of your PDF documents to evaluate that OCR software would struggle to extract.

We can give you relevant information and relevant suggestions. Like agent-specific alerts.

‍

Want to work with Harsh and the other humans of Prodigal as we pioneer consumer finance intelligence with the latest AI? We're hiring!