Technology
The Architecture
- 01
Mechanistic World Model
Fixed code turns the raw input into the features experts already use. Nothing here is trained.
- 02
Classifier Swarm
Small classifiers, each recognizing one pattern, together place the new example next to the past examples it resembles most.
- 03
Knowledge Universe
Every example the system has ever seen is kept here, with what it was, how it ended, and the record to prove it.
- 04
The Ether
For each action the system could take, it looks up how that same action ended in the most similar past examples.
- 05
Regional Skills
Each region of the map has its own skills: rules that remove forbidden options, and rankings built on the evidence from past examples. They change only offline, after passing sealed tests.
- 06
Frozen Language Model
It turns the finished decision into words a person can read, and it is never retrained.
Inside the system
The Knowledge Universe
This is how Ammonix stores what it knows. Each past example is a point on this map, and similar examples sit close together. A new example is shelved the way a librarian shelves a book, next to the ones most like it, and the outcomes of those neighbors are the evidence behind the decision.
Digital twins
Experience From Simulation
Many organizations have deep expertise but few recorded examples. Ammonix builds the missing experience by creating digital twins through modeling of the relevant processes: starting from a known answer, a generative model writes realistic examples of it, the way an examiner writes questions for an answer they already know.
This works because the model is never asked to answer a live example, where an invented answer could not be checked. It only writes training examples, and its mistakes show up as visible patterns in a dataset, measurable against real examples and correctable. Real examples enter the same knowledge universe as they happen and override the simulation wherever reality disagrees.