Technical overview.
Both products, in outline. What our systems guarantee and how we prove it in an engagement. Architecture and mechanism detail are shared in conversation, under NDA.
The Graph, in outline.
A governed longitudinal-state platform. It holds a long-run record of an individual (training, recovery, habits, self-reports) and draws conclusions that are true of that person, with four properties engineered in: individually true, accountable to their evidence, self-governed by the person the record describes, and honest by construction. Patent-pending.
It is domain-general: the vocabulary is configuration, the governance is the product. It runs in production today inside a live consumer deployment.
How records qualify, how conclusions are verified, and how erasure is enforced are the substance of the platform. None of that is published here; it is walked through, live, in conversation.
The Brain, in outline.
A governed development environment for AI-assisted engineering. The agent proposes; the Brain enforces: standing rules the agent cannot ignore, gates that can refuse an action outright, completion that must be verified before it may be claimed, memory that carries discipline from session to session, and continuous measurement of the whole apparatus.
The architecture, the enforcement model, and the tooling inventory are not published. What an engagement gets is better: the system itself, run against your codebase, with its measurements on the table.
How an engagement verifies us.
Every engagement starts with an acceptance procedure that runs locally, with no API key and no metered spend: you prove the system is installed, wired, and enforcing, with guards demonstrated against planted defects and verification judged by results rather than appearances, before relying on anything it says.
From there, the measurements do the talking: what fired, what it caught, what it cost, on your code and your records, not ours.