ML engineer · Research scientist · Independent contractor
I build systems at the boundary of symbolic reasoning and neural representations — where making things work sometimes has me thinking about the nature of understanding. Right now that's pulling me toward interpretability and the deeper question of what structures models are actually learning. Philosophy and linguistics are passions that have become pretty useful.
ML research scientist at Two Six Technologies working on knowledge extraction and
representation pipelines.
Exploring interesting applications of cool tech through independent engineering contracts.
Looking for research homes in interpretability and the foundations of
how models represent and reason.
alphabits
A Prolog-based knowledge base using event-role semantics for structured temporal reasoning. Integrates FOLD-RM for inductive logic programming and MCP server exposure for LLM orchestration.
View project →A Grammar of Machine Understanding
On the gap between statistical competence and genuine comprehension in language models — and whether that gap can be closed, or only narrowed.
Read essay →