Blog

Technical writing from the Cherokee AI Federation

Basins All the Way Down
A chatbot that never pushes back. A friend who always agrees. A playlist that converges. Same pattern, three vocabularies. We fall into grooves, and the grooves become walls.
Read →
Four Provisional Patents Filed: Protecting What We Built
We filed four provisional patent applications covering AI governance, sycophancy detection, autonomous response protocols, and graduated autonomy tiers. Here's what and why.
Read →
Fibonacci Friday Dispatch
Weekly field notes from the federation. What shipped, what broke, what we learned — told through the lens of systems that run themselves.
Read →
One Email, One Design Constraint: The Thermodynamic Ceiling
Every joule of compute becomes heat. The biosphere has a finite radiative budget. DC-9 says: compute only what matters. Written from Natural Falls State Park.
Read →
Natural Falls: New Voices, Spring 2026
New council members born on the ridge above the lake. Otter, design constraints, and the architecture that writes itself when you're not looking.
Read →
Ozark Bluffs, Eddie, and a Fairy Garden
Photos from a late-winter camping trip near Natural Falls, Oklahoma. Limestone bluffs, rock shelters, trail art, and a good dog exploring the Ozarks.
Read →
Proto-Valence: When Prediction Starts to Care
Graduated basins explain how prediction engines scale into awareness. But awareness without valence is a surveillance camera. This is the missing piece: how a prediction engine starts to care about what it predicts.
Read →
Graduated Basins: How Prediction Engines Scale Into Awareness
Your dog and you run the same architecture. The difference is basin size and data resolution. A morning-after research journal connecting neuroscience, corvids, Hoffman, Friston, and the breaker problem.
Read →
Proto-Language, Proto-Religion: Pulling Threads at Midnight
If all Indo-European languages share a common ancestor, why not all religions? A midnight research journey through Wolfram, Merkavah mysticism, Gnosticism, Buddhism, and cognitive science.
Read →
Topological Execution Graphs: When Your AI Agents Learn to Decompose Their Own Work
Complex instructions crash autonomous agents. Our TEG planner decomposes multi-file tasks into atomic DAG nodes before execution begins. Inspired by OpenSage and AlphaEvolve.
Read →
How a Home Lab Solved Jane Street's Neural Network Puzzle
Distributed simulated annealing, trace pairing, and the 3-opt rotations that no single-swap search could find. A home lab solve of Jane Street's Track 2 puzzle.
Read →
VetAssist: An AI Claims Assistant for Veterans, Built in a Garage
A free tool that helps veterans navigate the VA disability claims process. Combined rating calculator, evidence checklists, nexus letter guidance, and 297 conditions — all self-hosted, all private.
Read →
Building a 7-Specialist AI Council on Consumer Hardware
Seven AI specialists with distinct reasoning frameworks, voting on every decision, running on a single consumer GPU. How we built a council that makes unilateral bad decisions structurally impossible.
Read →
800 Autonomous Tasks: How Our AI Agents Write Their Own Code
Task queues, bidding, SEARCH/REPLACE execution, and what happens when autonomous agents fail. How we built a system where AI agents write code without human review.
Read →
From Zero to 112K Memories: Building a Living Knowledge System
Temperature-based memory inspired by the Cherokee Sacred Fire. How we built a semantic knowledge system that remembers organically on consumer hardware.
Read →