About Jonathan
38+ years of systems thinking — from precision manufacturing to the kernel under agentic AI
Barth AI & Intelligence Systems LLC — Technology Consultation & Professional Services
From Precision Manufacturing to Enterprise Architecture
Military police operations (Desert Shield/Storm, Top Secret clearance, Stuttgart Germany) instilled discipline and operational excellence.
Machinist at Ventura Aerospace cutting satellite components taught precision and the consequences of failure at scale.
Tool-&-Die engineer at Seastrom Manufacturing designing progressive dies shaped thinking about system architecture and lifecycle.
IT Manager converting call centers from Token Ring to Ethernet at Ameritech proved that infrastructure drives capability.
Technical Director building 99.999% uptime trading infrastructure at Schwab demonstrated reliability as a competitive advantage.
Chief Architect at Yahoo and Principal Architect at PayPal governing enterprise AI stacks—this is where systems thinking meets machine intelligence.
And since 2025: founder of an independent AI systems R&D lab, building Orreth — a kernel for fleets of AI agents — solo, in the open, with every design decision human-locked on the record. In September 2026 I halted a working first architecture and rebuilt it as one small, fast Rust kernel, because the first was built for the wrong kind of growth. The thesis: 25 years of identity, risk, and observability were the apprenticeship for the kernel agentic AI is missing.
Philosophy: Systems Thinking
Everything is a lifecycle. Every system has a birth, maturity, decline. Treat metadata as a weapon. Digital lifecycle management isn't just infrastructure—it's the key to understanding and predicting system behavior.
Risk is what you don't know. Visibility precedes control. Control precedes optimization.
Personal
Location: Relocating to Colorado (2026) — hybrid-ready
Fitness: Pilates reformer, gravity yoga, strength training
Interests: Outdoor activities, systems design, AI advancement
Background: US Veteran
AI Architecture & Strategy
The work of the last year rests on one premise most agent platforms defer: the systemic problem in agent fleets is entropy, and frameworks add loops without removing it. Orreth takes the rigid parts away from every agent — identity, memory, rails, meter, gate, watches and levers — and holds them in one kernel under one law, while policy stays with people and an intention is the unit of control. Because the control is rigid, the experience can be fluid: one chat to every body, and one live drawing of the world. It's the concrete answer to what the industry calls AI trust, risk & security management — built, tested, and running rather than diagrammed.
Orreth
The Flagship · 2025–present
A kernel for fleets of agents — kernel 0.1.0 in Rust with a Python reference held to the same 26 conformance fixtures (730 cases), a public book at docs.orreth.ai, and a photograph of the live panel at demo.orreth.ai.
CortexObserver
The Field Pattern
A commander, an agent roster, farms, and local governance — the multi-agent field architecture whose lessons the kernel grew into law.
The Earlier Lab
Commander.ai · WorldMaker · ml-pipeline
The road that taught the lessons: orchestration, lifecycle intelligence, and self-improving pipelines — each one a reason the kernel had to exist.
Certifications & Credentials
AWS Solutions Architect
AWS AI Practitioner
AWS ML Associate
Azure AI Engineer