I build full-stack products and applied AI systems that survive real use.
I work across frontend, backend, data flow, applied AI integration, and the reliability layer that makes software usable after real constraints show up.
I am based in Newcastle, Australia, and I work with teams locally and remotely. The work is usually one concrete product surface, system slice, or workflow at a time.
What I optimize for
I am less interested in looking cutting-edge than in making a system legible, reliable, and worth operating.
Production-minded delivery
I build AI workflows, product systems, and automation infrastructure with tests, observability, and explicit failure boundaries instead of demo-only optimism.
Full-stack implementation
I work across workflow design, backend systems, APIs, frontend surfaces, and the reliability layer that keeps the whole thing usable.
Governed AI focus
The work I care about most is the gap between agent hype and systems that can survive real operational use.
How I think about the work
The useful part of AI infrastructure is not that an agent can act. It is that the surrounding system makes those actions bounded, explainable, and operationally sane.
That is why my work leans toward product clarity, validation layers, approval paths, and production hardening. The goal is not to make a business feel futuristic. The goal is to remove brittle manual drag without introducing a worse failure mode.
Operating principles
- • Narrow workflows beat vague autonomy claims.
- • Approval points and validation belong in the execution path.
- • A useful system is observable, testable, and reviewable.
- • The point is not more AI activity. The point is less operational drag.
Systems behind the positioning
The brand direction is backed by real systems work, not just copy. These projects are part of the proof.
Portarium
The control-plane direction behind my work on governed AI systems: policy, approvals, orchestration, and evidence around actions that matter.
OperatorOpenClaw
The operator/runtime direction I am prototyping for bounded business workflows, approvals, and auditable execution.
Operational proofVibeCord
A real product story with users, deployment complexity, and post-mortem lessons about what breaks when systems are shipped faster than they are governed.
Background
The common thread is building and shipping systems end to end, then learning where they break under real use.
AI systems and workflow engineering
Building agent systems, automation pipelines, content infrastructure, and the reliability layer around them.
Industry-backed AI engineering work
Built an AI agent backend for Appaca through a university-industry project, which sharpened the gap between prototypes and production systems.
Full-stack product delivery
Shipping web applications, backend systems, tooling, and internal workflow software across solo projects and client-style builds.
If you need one product or system slice shipped properly, start there.
The best work usually begins with one concrete product surface, one real bottleneck, and a willingness to make the system clearer before making it bigger.