Konrad Reczko
@reczko_konrad
Software Engineer @ Software Mansion working on TypeGPU ๐
341 Following 7.6K Followers
TypeGPU + ruNNtime + Jev @typesafeai is a very fun combo :D
ruNNtime gives me efficient local inference, TypeGPU lets inference and rendering share GPU resources directly with zero copy. Thatโs 3 separate NN inferences plus rendering, all happening in realtime
Since we control the pipeline, Jev can just sit in the middle and add the semantic bit.
camera + mic โ Moonshine + YOLO26 + DepthART โ Jev โ lights, shadows and bloom
Show more
Thank you all for the kind words under the demo I posted :D
The demo is now officially public and open source for you to play with, modify and build on top of. Feel free to tag me if you make something cool with it :)
Also thank you @iwoplaza for participating in my silly video haha
You can run it like any other TypeGPU example here:
The source code lives in the TypeGPU monorepo:
The repacked weights used by the example live here:
Show more
Depth-aware light injection in TypeGPU
I got a 448x448 monocular depth model down to ~8 ms on my M4 Pro across ~250 dispatches, which is fast enough to use in realtime :D
Since the inference is written directly in TypeGPU, I can just feed the depth buffer straight into the lighting pass. It never has to leave the GPU or go through any extra synchronization/interop step
Inference, lighting and draw all go through the same command encoder.
Show more
Writing games in TypeScript from scratch sounds crazy, right? Maybe, but I think it's not only possible but also performant, nice for DX and works both on Web and React Native
To prove that point i created Bone Tide, it's a game that was written purely in TypeGPU. No engine, both game logic and shaders written in pure TypeScript
You can try the web version right now:
On mobile, I recommend adding it to your homescreen as a PWA
Show more
I love when a stack this absurdly layered somehow just works
Iโm running custom segmentation inference on VisionCamera frames in React Native, using TypeGPU TypeScript shaders through react-native-webgpu. The segmentation itself runs on the GPU using zero copy hardware frames, while the frame orchestration and control logic stay on the UI thread via react-native-worklets.
And somehow SimCam can still turn a region behind the iOS Simulator into the camera feed or seamlessly swap back to my laptop camera.
Very silly, very useful, I love it :D
Show more
