I'm currently working on backend engineering at Mirror Mirror AI, mainly on the backend for AI image generation, likeness licensing, and content provenance, while spending much of my time exploring LLM inference systems, which I believe are a technical foundation for unlocking new possibilities for products and businesses.

Lately I've been especially fascinated by how models actually get run, from GPUs, runtimes, and distributed systems all the way to the LLM serving engine. I like to take apart open-source projects like SGLang and vLLM, reading the code and running experiments, peeling back my understanding of AI infrastructure layer by layer.

I love to build things, and I love understanding things under the hood.

When I'm not working, I'm usually lifting, diving, dabbling in investing, or falling down some new rabbit hole.