I’m a machine learning researcher and engineer interested in how intelligence works, not just how to scale. Most of my work is around language, vision, mathematical, and physics informed models, with a particular interest in long horizon reasoning, and multimodal sequence modeling. I build, deploy, and evaluate these systems on distributed home clusters and edge hardware, such as; Google Coral Tensor Processing Units (TPUs) on Raspberry Pi.

I’m particularly drawn to the ideas I believe will shape the field’s next chapter: Energy Based Transformers (EBTs), Diffusion Language Models, and Photonic Computing, Cybernetics, and the broader questions around mind, perception, and computation.

Beyond my own publications & projects, I contribute to open source and believe the most meaningful advances emerge through collaborative research. ᝰ✍︎

“ Most of human knowledge is actually not language so those systems [LLMs] can never reach human level intelligence; unless you change the architecture. ” - Yann LeCun

“ Large language models can talk endlessly because they are trained on huge bodies of knowledge; but genuine intelligence is “the ability to create knowledge; spot a problem, invent a solution, test it, and improve it as humans do. ” - David Deutsch