Blog

Notes on intelligent systems, mathematics, and building strange things.

Why Do LLMs Think They Are Human?

MAR 05, 2026

A deep dive into self-anthropomorphism in LLMs.

AI SAFETY POST-TRAINING

ClaudeOS: How I Learned to Stop Worrying and Love the Slop

FEB 11, 2026

A reflection on the current abilities of agentic coding and the story of vibecoding a functioning OS that can play DOOM, in TWO DAYS.

AI OS VIBECODING