←back to thread

Ask HN: What Are You Working On? (June 2025)

439 points david927 | 2 comments | 29 Jun 25 20:21 UTC | HN request time: 0.711s | source

What are you working on? Any new ideas which you're thinking about?
Show context
jesse__ ◴[29 Jun 25 22:01 UTC] No.44416890[source]▶
I've been working on a 3D voxel-based game engine for like 10 years in my spare time. The most recent big job has been to port the world gen and editor to the GPU, which has had some pretty cute knock-on effects. The most interesting is you can hot-reload the world gen shaders and out pop your changes on the screen, like a voxel version of shadertoy. https://github.com/scallyw4g/bonsai

I also wrote a metaprogramming language which generates a lot of the editor UI for the engine. It's a bespoke C parser that supports a small subset of C++, which is exposed to the user through a 'scripting-like' language you embed directly in your source files. I wrote it as a replacement for C++ templates and in my completely unbiased opinion it is WAY better.

https://github.com/scallyw4g/poof

replies(6): >>44418041 #>>44419836 #>>44424357 #>>44433687 #>>44434050 #>>44456447 #
1. mentos ◴[01 Jul 25 13:24 UTC] No.44433687[source]▶
Curious if you have started using LLMs to speed up any of your development yet?
replies(1): >>44435892 #
2. jesse__ ◴[01 Jul 25 16:58 UTC] No.44435892[source]▶
No. The bottleneck for me is not the speed at which I type code, which is basically what LLMs accelerate. I've been thinking about trying again, but at this point the context windows are FAR too small to feed my entire project too, and it just seems like a giant pain to have to babysit that.

Additionally, the bottleneck for me is typically writing graphics code, and LLMs are hilariously bad at that.