Hi HN! We're really excited to share Apres with all of you. Our goal is to make developing and maintaining AI models easier by reducing the burden of data creation/management. Give us a try (your first 2,500 requests are free)! And let us know if you have a use case we don't support yet.

Looks great. Does this need a bunch of unlabeled training data to begin with, and then suggests datapoints to label, or can it also suggest labels for which data is needed next?