I build industry solutions on the Microsoft stack.
Twenty years from .NET to agentic AI. Product engineer and solution architect for property, construction and government. Currently building BTR OS.
Systems people rely on.
Built end to end, shipped to production, and run. This is the work that proves the method.
BTR OS. The operating system for Build-to-Rent.
Six operational modules, a resident app, a governed AI agent surface and one audit trail, on Azure and .NET. I designed it, built it with governed agents, and run it.
Explore BTR OSA safety-critical agent, governed from day one.
Workforce access and compliance for a major civil program, with an AI agent working the toolbox-and-hazard-report flow: proof of concept delivered in one week, now in design and build for production.
The phone is the key. The stack is the staff.
Mobile access control and billing for unstaffed gyms, delivered full-stack across backend, web and mobile: members join, pay and open the door without a human on site.
A Power Platform Centre of Excellence, at federal grade.
A governed low-code delivery capability inside a federal science agency, built and run at full ISM and IRAP compliance: citizen developers ship, governance holds.
Enterprise modernisation at construction scale.
Real-time construction reporting, finance APIs and contractor portals across thousands of subcontractors for Tier 1 builders, plus HSE safety applications and a SharePoint Online migration that moved the business.
Backend web services and reporting portals in C# and .NET.
Microsoft PPM and SharePoint for government and enterprise, plus the EPMonDemand ISV product.
Agents write the code. ArkOS decides what ships.
Every change I ship is grounded in a spec, checked by gates, and recorded on an audit trail. The framework is public.
- 01
A spec is drafted and approved: problem, EARS requirements, acceptance criteria.
- 02
An issue anchors the change; a feature branch carries it.
- 03
Agents implement against requirement IDs, with humans on judgment calls.
- 04
Acceptance checks run: greps, builds, manual review against each REQ.
- 05
The PR merges through the gates. The audit trail keeps the whole story.
Twenty years on Microsoft.
The stack is deliberately boring, so the new thing is what ships on top.
Microsoft Azure
Runs everything I ship
LIVE.NET
The backbone since 2006
LIVEMicrosoft Entra
Identity on every system
LIVEPower Platform
Governed low-code at federal grade
LIVEMicrosoft 365
Where the users already are
LIVEAzure AI Foundry
Powering the agents
IN BUILDFrom .NET services to agentic delivery.
Product engineer & solution architect, Canberra & Sydney.
Backend web services and reporting portals in C# and .NET.
Microsoft PPM and SharePoint for government and enterprise, plus the EPMonDemand ISV product.
SharePoint Online migration, construction finance APIs and HSE safety applications for a Tier 1 operator.
Enterprise modernisation: real-time construction reporting and contractor portals across thousands of subcontractors.
A Power Platform Centre of Excellence inside a federal agency, at full ISM and IRAP compliance.
Workforce access and compliance for a major civil program, with a safety-critical AI agent: proof of concept delivered in one week, now in design and build for production.
Mobile access control and billing for unstaffed gyms: phone as key, full-stack delivery across backend, web and mobile.
The operating system for Build-to-Rent. Specialist practice and product architecture for launch, operations, integration and intelligence.
Thinking, published.
Analysis from inside the work: markets, governance and agents.
Why Australian Build-to-Rent needs its own operating software
Biometrics, privacy and the audit trail: access control under the Privacy Act
What governed AI agents mean for building operations
Three ways to engage.
One person does the work you brief. I reply within a business day.
Embedded contract
A product engineer and solution architect inside your team: delivery, architecture and the judgment calls in between.
Solution delivery
A defined system, built end to end: spec, build, ship, operate, with governance on from the first commit.
Advisory
Architecture reviews, agentic adoption and governance: how to get agent throughput without agent drift.