AI software engineer working on inference engineering, model optimization, and edge AI for robotics: making capable models run fast on real hardware.
/\ /\
/ \__/ \
/ O O \
| ︶ |
| \____/ |
\ ~~~~~~ /
\________/
vla.cpp: Vision-Language-Action models on local hardware, in the spirit ofllama.cpp.vla.simd: efficient CPU inference for language-conditioned manipulation - SIMD micro-kernels (AVX2 / NEON) running 30+ actions per second on a Raspberry Pi 5.- REX: neural network Representation EXchange - a compiler converts common models between machine learning frameworks.
- ReadItDown native markdown viewer and editor in Linux/Windows/macOS.
- I optimize deep neural networks for AI accelerators: VLM · VLA · LLM.