A pseudo operating system for local and cloud compute (Opensourcing on 5000th star)

CUDA kernels in any language supported by LLVM

Revives the last functional CUDA deep-learning stack on macOS High Sierra. NVIDIA CUDA 10.2 + cuDNN 7.6.5 + PyTorch 1.7.0 fully working on GTX Pascal GPUs — a discontinued ecosystem resurrected.

A ray-tracer/path-tracer implemented in JavaScript using the WebGL 2 API

.NET binding for wgpu-native

The .NET wrapper of the Cocaine C library. Cocaine is a multi-platform C library that can be used to accelerate large workloads/big data/anything really with the power of a GPU with ease.

Browser-based Finite Element Analysis solver showcasing WebGPU compute and Rust → WebAssembly numerical pipelines.

Silicon Fingerprinting & Hardware OSINT Engine leveraging GPU-accelerated telemetry analysis and microscopic clock-skew variance. Built with Rust, C, and HIP/ROCm.

Cocaine is a multi-platform C library that can be used to accelerate large workloads/big data/anything really with the power of a GPU with ease. A .NET wrapper is available in the link below.

GPU Implementation of Mandelbrot Fractal Generator with Benchmarking

Voxel engine from scratch in Swift + Metal for Apple Silicon

Autonomous ML research agent skill (Agent Skills format). Designs experiments, deploys to Modal GPUs, tracks budget, iterates autonomously.

GPU based mean filed/rate theory/cluster dynamic packages for SiChuan university.

GPU Implementation of Newton Fractal Generator with Benchmarking

An HLSL implementation of a thread group wide bitonic sort for GPU

🏛️ The Power Economy | Sovereign Infrastructure & BTM AI Assets. Featuring behindthemeterpower.com & behindthemeterai.com. 📧 ceallo6@proton.me

Common Lisp minimal GPU code example

Conceptual path finding in high-dimensional embedding space via Grassmann manifold geometry. Replaces cosine similarity with tangent space distance — paths follow domain-coherent lanes instead of converging to vocabulary hubs. GPU-accelerated Julia compute worker

(Attempted) Starter execution layer for CUDA devices

ComputeScan analyzes tfplan.json to catch GPU oversizing, autoscaling misconfigurations, tag drift, and high-risk idle patterns. Instantly, offline, and with zero setup. Designed for AI/ML infra teams and fully aligned with the GuardSuite governance engine.