Constant-complexity, deterministic, very fast memory allocator (heap) for hard real-time high-integrity embedded systems. Allocation takes ≈120 cycles @ RP2350 irrespective of heap usage. There is …

A path-tracer for the original Quake game

Lightweight coding agent that runs in your terminal

Surfel Radiance Cascades Diffuse Global Illumination (SRC-DGI)

HomebrewGPU is a simple ray tracing GPU built on FPGA, featuring basic ray–primitive intersection, BVH traversal, shadowing, reflection, and refraction. I created it to learn Verilog programming, a…

Voxel rendering experiments

State of the art sorting and segmented sorting, including OneSweep. Implemented in CUDA, D3D12, and Unity style compute shaders. Theoretically portable to all wave/warp/subgroup sizes.

Single-header dependency-free BVH construction and traversal library.

Library for collision detection between two convex shapes

A project for testing rigid body solver algorithms

GPT4All: Run Local LLMs on Any Device. Open-source and available for commercial use.

Excalibur Hash is a high-speed hash map and hash set, ideal for performance-critical uses like video games

Focus on prompting and generating

LLMs build upon Evol Insturct: WizardLM, WizardCoder, WizardMath

build-once run-anywhere c library

Distribute and run LLMs with a single file.

An OpenGL sample that demonstrates Volumetric Fog using a frustum-aligned voxel grid and compute shaders.

Code and documentation to train Stanford's Alpaca models, and generate the data.

Inference code for Llama models

LLM inference in C/C++

Multi-channel signed distance field generator

Build android apps without any java, entirely in C and Make

GPU Raytracer from scratch in C++/CUDA

Voxel ReSTIR a ReSTIR implementation for the WrldTmpl8 framework.

An implementation of NVIDIA's paper "Efficient Incoherent Ray Traversal on GPUs Through Compressed Wide BVHs"

Audio playback and capture library written in C, in a single source file.

Coupled simulation of FEM continuum media and articulated rigid bodies

With adaptive merging we reduce computation time in rigid body simulations.