OK here is my big announcement. For the past couple years I've been working on a rapid prototyping and development platform for real time rendering.
TL;DR - use nodes in a node graph to string together compute shaders, ray gen shaders, draw calls, etc. View it in a viewer that supports hot reloading. When you are done, generate code that would pass a code review. (Currently only DX12 code gen is public. More coming in future!)
https://github.com/electronicarts/gigi
GitHub - electronicarts/gigi: A framework for rapid prototyping and development of real-time rendering techniques.

A framework for rapid prototyping and development of real-time rendering techniques. - electronicarts/gigi

GitHub
@demofox looks like a wish come true! If I’m ever using windows again I can see myself getting lost in it haha.
@photex the plan (not sure when) is to add vulkan generation, and get the viewer (interpreter) able to run vulkan as well, then hopefully it should be able to be mac and linux too. But yeah, not here today unfortunately!
A driving force for this though is that a surprising amount of ML researchers use mac and linux, so who knows, maybe that'll help it happen sooner.
@demofox it is what it is. Work and life just happen to have me working on Linux currently and I gave my personal laptop to my son earlier this year which is mostly being used to play Elden Ring at this point haha.
But yeah, our entire ML and perception teams are basically Mac users with one or two using Ubuntu. Everyone else is on one Linux or another with the exception of some management folks who were born and will die in Office 365.
@demofox so one of the things that bothers me immensely about AI code assistants is just how off the mark they are. (Typing isn’t my problem, it’s designing and modeling).
If you look at something like simulink (despite its many problematic aspects), it lets you build models for complex systems and then generate code or firmware and visualize things. This is what I’m picturing here except for rendering pipelines.
Something that is domain specific and is a toolbox to let you design things a rung up the ladder and plugs into a broader codebase etc.
It’s really really exciting! I hope it catches on.
@photex agreed and I hope so too.