I build bounded proof surfaces for AI governance at the point where system action becomes consequence.
Where does the system physically stop?
The work focuses on execution-boundary control, admissibility, refusal, authority, receipts, and replay.
claim → evidence object → inspection path → claim limit
Public GitHub is an inspection surface, not full architecture disclosure.
- start-here — canonical entry point
- commit-gate-core — bounded proof that mutation requires a valid DecisionRecord
- inspection-surface — public research and provenance surface
Each public repository supports one local claim with an inspectable evidence object and an explicit stopping point.
Unless explicitly stated, these artefacts do not prove:
- production readiness
- compliance or certification
- enterprise deployment
- adoption or standardisation
- path-universal governance
- full architecture disclosure
Architecture, methods, and system designs across this profile are the original work of Ricky Dean Jones, unless otherwise stated.
LinkedIn: linkedin.com/in/ricky-jones-1b745474
Status: active research and engineering work.