Getting Started with Rivet 1.5 

Great work guys! One thing would be cool to have. Deletion of the nodes by clicking delete or backspace button. It's not very comfortable to open context menu every time you want to do that. But in any case, it's supercool project. Thanks a lot for your work.

I'm loving this so much. I'm new to coding and AI and this app give me all the internal wigglys of wanting to explore these two worlds!

I'm finding Rivet really great for putting together complex sequences of prompts. The loop tool is a bit tricky to get your head around, and it would be great to be able to plug in other llms (local ones for example, or autogen), but this is still a brilliant tool.

Thanks for the nice intro to Rivet!

Very clear. It helps a lot to get started. Thanks

Cool work, why not make a library of graphs - and "prompt search" to find those that do something similar to what you want to do?

How does the Critiquer graph function? That's sort of helpful in order to figure out how this works to revise a document.

It would be very useful to have a very simple example of how to turn this workflow that you created into a working web page with a form for the inputs, or even just how to run this from a script outside of Rivet.

can you please post how to use this in our local production

Also, in the example above, how did you add the extra loops to review the information and to revise and refine it? Why did you need to do it as two sub graphs versus one that simply reviewed and rewrote it with a tighter specification? Is there some thought process there that we need to know about and if so, could you walk through how to do that so that we could start to model things like that ourselves? For example, let's assume you're trying to create an application that has five inputs and a dozen different outputs that come out of open AI based on various prompts that may also use the variables gathered in the inputs. It'd be great if you could review and refine and revise the output prompts each time. Can you walk us through how to create a sub graph that reviewed and refined each prompt o so that we only had to write this logic once, but we could use it throughout our entire process for all 10 prompts to end up with a more refined output than we would have received had we just run through the whole chat sequence, one time? Showing us how we could we create this in rivet would be super helpful.

Absolutely brilliant 🥹

Just a quick suggestion, can you perhaps "automatically" make "wrapper" for the autogen api and for the functions? I guess doing that would make a huge leap!

Thank you!

Hi, THX for this,! It would be very helpful to se how the other two subgraphs worked for someone like me who is completly new to this. Can I find that somewhere?

Nice.....A cool tool....

You rushed the end of the video and im stuck trying to figure out how you did it.

So...what do I do with this now?

No Lm studio support? That's an abomination!