I like this, are you able to share any details on how you built this? Or you probably have some bigger plans for this.

That is cool, do you have a timeline set out for this?

I'll likely start by creating a web API service similar to what I did for gpt-2-simple, except more efficient: https://github.com/minimaxir/gpt-2-cloud-run

The next step is architecting an infrastructure for scalable generation; that depends on a few fixes for both aitextgen and the base Transformers. No ETA.

reminds me a bit to: https://github.com/eldarion/eldarion-ajax

So no one really knows? I guess going with your gut would be the best answer here.

Paraphrasing Felix Dennis: Barely passable ideas with good marketing and timing can work if the execution is solid.

This also holds up to other research concluding timing is the biggest factor.

I've seen a number of solid startup teams cash in on recycling good ideas from the past (m&a shutdown) and execute well to satiate similar, modern needs. Novelty can be risky (Pebble / Apple Watch) because the latest entrant can learn and improve on what came before.

Not mine:

http://drum.jupo.org/

blog post here:

http://blog.jupo.org/2013/04/30/building-social-apps-with-me...

You could try:

http://gettingstartedwithdjango.com/

or

https://godjango.com/

The Restangular url is here: https://github.com/mgonto/restangular

Built with django running osqa. My question is, how can I grow a community that would find this useful?

Any suggestions welcomed.