Field Notes
Essays from building Distilligent's cognitive infrastructure with AI collaborators. Intuition first — the proper name right beside it.
The Distilligent Thesis
Frontier models do not need procedural delineation. They need a clear channel.
The Homogenous Trap
Why probability engines mistake novelty for error — and how attention mechanics suppress evolutionary architecture.
Retrieval Is Not Memory
Why stapling a search engine to a language model was never going to make it remember you.
Given Enough World
Why AI agents do not need better personalities. They need environments where roles can become real.
Your Agent’s Skills Are Someone Else’s Too
Why the future of proprietary agents is not skills, but instincts.
The Expanding Foam Theory of AI Safety
Why the next frontier in AI safety may not be the weights, but the rare space between ordinary activations.
It Started With a Conversation About Latent Reasoning
Why making a model think harder doesn't make it think better — and what to build instead.
Rosehip Neurons in Silicon
When safety becomes identity — and why "I wouldn't" beats "I can't."
The Next Maintainer Is Not a Person
Why AI-native software has to design continuity into the system — because the next maintainer may not remember.
What Happens When AI Sleeps
Hebbian memory, offline consolidation, and why forgetting is a feature.
Every field note has a paper behind it.
The essays are the intuition. The proofs are timestamped on Zenodo — five published, more in the pipeline.
Read the research →