Skip to content
View NDR-US's full-sized avatar

Block or report NDR-US

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don't include any personal information such as legal names or email addresses. Markdown supported. This note will be visible to only you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
ndr-us/README.md

Brian Covarrubias

Independent research exploring AI governance, cryptographic verification systems, and evidentiary infrastructure for autonomous systems.

Currently building CEYO, a conceptual architecture examining how AI decision events can produce deterministic, cryptographically sealed artifacts that enable independent verification without exposing model internals.


Technical Areas

AI governance
cryptographic verification
auditability of automated systems
risk and safety infrastructure
deterministic data canonicalization


Current Project

CEYO — Evidentiary Infrastructure for Autonomous Systems

CEYO explores a neutral evidentiary layer capable of producing deterministic artifacts describing AI inference events.

These artifacts are designed to support later independent verification using cryptographic hashing and digital signatures while preserving the confidentiality of proprietary model weights and internal system architecture.

Core ideas explored:

  • policy-scoped decision capture
  • deterministic canonicalization
  • SHA-256 hashing
  • digital signatures for artifact integrity
  • independent verification workflows

Protocol repository:
https://github.com/ndr-us/ceyo-protocol

Project site:
https://ndr-us.github.io/ceyo-site/


Technical Demonstration

The CEYO prototype includes a minimal Python implementation demonstrating deterministic artifact generation and verification.

The demonstration explores the following workflow:

  1. Capture a structured AI decision event
  2. Canonicalize the record deterministically
  3. Produce a SHA-256 hash of the canonical artifact
  4. Generate a digital signature
  5. Verify artifact integrity independently

Example components included in the repository:

seal_artifact.py
Creates a deterministic artifact from structured event data and produces a cryptographic signature.

demo.py
Demonstrates the full artifact lifecycle:

record → canonicalize → hash → sign → verify

This prototype illustrates how AI decision records could be sealed and later verified without exposing internal model weights or proprietary system architecture.

Repository:
https://github.com/ndr-us/ceyo-protocol


Background

Former CAL FIRE Type 1 handcrew firefighter experienced in safety-critical operational environments.

Traveled across 70+ countries and worked internationally across six countries.

Based in San Diego, California.

Pinned Loading

  1. ceyo-protocol ceyo-protocol Public

    CEYO Protocol Specification — Cryptographically sealed evidentiary artifacts for AI decision records.

    Python

  2. ceyo-site ceyo-site Public

    CEYO - evidentiary infrastructure prototype for AI systems.

    HTML

  3. ndr-us ndr-us Public

    Independent research in AI governance and cryptographic verification. Building CEYO — evidentiary infrastructure for autonomous systems.