Systems programmer. I build autonomous & ML-driven systems, speech-recognition pipelines, operating systems & GPGPU compute systems - Mostly From Scratch, also do integration work.
Currently · groundstation (autonomous drone flight management) and sttserv (multi-model speech recognition).
groundstation
C++ · ML-driven autonomous drone flight management via a natural-voice command interface.
sttserv
C++ / Python · speech-recognition (ASR) abstraction over Sherpa-ONNX and whisper.cpp; English-first, Hebrew-capable.
| project | stack | |
|---|---|---|
| UEFI bootloader + kernel (see also x86-smp-boot) | C |
|
| Windows DRAM tail-latency via channel-replicated, hedged memory reads | C++ |
|
| real-time 2D fluid (GPU Gems ch. 38, OpenGL compute shaders) | C++ |
|
| GPU compute-shader raytracer, cross-platform (Windows/Linux) tech demo | C++ |
|
| cross-platform C/C++ utility library: allocators, hashing, RNG, ABI-stable C API | C++ |
Full stack
Languages: C++, C, Python, x86-64 Assembly, GLSL, C#
GPU / Graphics: OpenGL compute, GLSL, Vulkan, SPIR-V, glbinding, GLFW, Dear ImGui, SDL3, shaderc, GLM
Systems / OS: UEFI (GNU-EFI), x86-64 boot, QEMU/OVMF, Win32, MemProcFS, miniaudio
ML / AI: llama.cpp, whisper.cpp, Sherpa-ONNX, faster-whisper, CTranslate2, Transformers/PyTorch, ONNX Runtime, NumPy
Robotics: PX4, Gazebo, OpenVINS, StellaVSLAM, GStreamer, OpenCV, Eigen
Build / Test: CMake, premake5, Ninja, Docker, GoogleTest, Google Benchmark, GitHub Actions