Writing

Research papers, highlighted essays, and other posts.

Research

• Measuring What Neural Networks Really LearnResearch
  A framework for estimating intrinsic dimensionality through geometric distillation. Published in arXiv 2512.00792.

Highlighted

• Snowball CriterionHighlighted
  On recursive improvement and intelligence.
• Why We're Alone: The Gradient Descent HypothesisHighlighted
  A Great Filter hypothesis on convergent optimization and extinction.
• Symbiosis: The Only Way ForwardHighlighted
  Humanity's path toward symbiosis with AI and transcendence.

All Posts

• Measuring What Neural Networks Really Learn
  Framework for estimating neural network intrinsic dimensionality via geometric distillation.
• Snowball Criterion
  A criterion for advanced intelligence: the ability to design a more intelligent successor.
• Why We're Alone: The Gradient Descent Hypothesis
  A Great Filter hypothesis on convergent optimization and extinction.
• Symbiosis: The Only Way Forward
  Toward a symbiotic future of biology and silicon.
• On the Sisyphean Nature of Life
  Reflections on meaning, Camus, and defiance of the absurd.
• A Graph-Based Programming Language
  Introducing Automa: a graph-based model for automata and state transitions.