I think there could be some cool generative UI stuff to build with it.

Based on certain actions, choose a set of components to render. Want to give that a go once I get a chance.

I’m building an eDiscovery tool that helps me categorize and classify artifacts for potential data loss, policy violations, legal violations etc.

I am trying to use it to categorize content, detect priority, and whether the conversation matches with the goals defined by the users.

[dead]

[dead]