Government does not fail at digital delivery because the technology is hard. It fails when legislation, policy intent, security obligation, and delivery reality are held by four different people who never share a document.
I hold all four. As Portfolio Digital Architect for the Government of Alberta, I set architectural direction across five ministries and translate statute and service need into architecture that teams can actually build against. My primary ministries are Affordability and Utilities, Energy and Minerals, and Indigenous Relations. I also support Treasury Board and Finance and Children and Family Services.
My work is deliberately close to delivery. I have taken an executive-priority system from requirements to production in six weeks, stood up new tribunal and case-management platforms on Alberta’s digital service platform, and shipped the AI playbooks and reference patterns that let other teams do the same thing without re-deriving the governance every time.
Portfolio and enterprise architecture, architecture review, decision records, trade-off analysis
AI delivery playbooks, RAG systems, Model Context Protocol knowledge services, AI-first governed delivery
Protected B & C hosting, OpenShift, identity, API contracts, information-protection patterns
Legislation-to-architecture translation, tribunal and case management, fixed-mandate launch dates
I work at the portfolio level, where the decisions are expensive to reverse.
Five ministries, one architectural direction. I am the primary architecture decision-maker for digital services in Affordability and Utilities, Energy and Minerals, and Indigenous Relations, and I also support Treasury Board and Finance and Children and Family Services.
AI is only useful in government if it is governable. My work makes AI-assisted delivery repeatable, auditable, and aligned to enterprise standards.
High-sensitivity environments do not tolerate late security thinking. I bring hosting, identity, and information-protection questions into the design before the first sprint.
Governance earns its place when it makes delivery faster.
Key accomplishments and the value they delivered, measured rather than described.
Executive-priority system: From requirements to production, including architecture, API contracts, deployment, and production readiness.
Tribunal and case-management platforms: Established on the Alberta Digital Service Platform and running in production.
Consolidated architectural direction: Five ministries aligned under a single architectural direction.
Enterprise AI acceleration: Three reusable enterprise AI assets adopted by multiple delivery teams.
Highest-sensitivity secure hosting: Supporting work on OpenShift hosting and secure integration patterns for sensitive systems.
Public registry, scratch to production: Built alone in two weeks of development time with an AI-assisted, specification-first workflow. The pilot succeeded.
Lead architect for an executive-priority tribunal program working to a fixed launch date. Shaping solution architecture and sequencing to hold that date without trading away security or governance posture.
STACK ADSP · Case-management architecture · Secure integration · Reusable component libraries
Established the architecture from scratch on Alberta Digital Service Platform services, brought it to a stable state, and took it to production.
STACK ADSP · Case-management architecture · Cloud-native services · Platform-service integration · CI/CD
Led end-to-end architecture under a six-week executive timeline: refined requirements, drafted the implementation plan, defined API contracts, and oversaw deployment.
STACK Azure · API contract design · Azure DevOps · CI/CD
Bootstrapped a public-facing registry from scratch to production in two weeks of development time, about 55 hours. I built it alone with an AI-assisted, specification-first workflow: agents produced the requirements and architecture documents, and the application was built from them under the AI delivery playbook.
STACK AI agents · Specification-first delivery · ADSP · GitHub Copilot
Built the assets that make AI-assisted delivery repeatable across government: an AI delivery playbook, an AI toolkit, and a Model Context Protocol architecture knowledge service deployed on ADSP.
STACK MCP · RAG · Agentic workflows · GitHub Copilot · ADSP
Architectural direction for systems that predate the current platform. Each engagement starts with a technical system overview and a pain-point analysis, then produces a recommended path and the architecture to execute it. Three run concurrently.
STACK AEM · GitHub Enterprise · Azure Boards · Azure Pipelines (YAML) · PowerShell automation
A vendor-neutral architecture framework for adopting enterprise AI in government, public-sector, and regulated organizations. It covers architecture, governance, and security, with compliance mappings and an AI-assisted delivery model.
STACK MkDocs · Markdown · GitHub Pages
Agents for GitHub Copilot and Claude Code that take an idea through ideation, architecture, and implementation, grounded in an organization’s own standards. Code agents work only inside approved work packages, and each package closes after an independent review.
STACK Python · GitHub Copilot · Claude Code · MCP
Principles I design by
| Credential | Issuer | Issued |
|---|---|---|
| Alberta AI Academy, Levels 1 to 3 | Government of Alberta | Aug 2026 |
| Micro-Certification: Welcome to ServiceNow | ServiceNow | Aug 2026 |
| Certified SAFe® 6 Advanced Scrum Master (expired Dec 2024) | Scaled Agile, Inc. | Nov 2023 |
| Certified SAFe® 6 Product Owner / Product Manager (expired Dec 2024) | Scaled Agile, Inc. | Nov 2023 |
| Certified SAFe® 6 Agilist (expired Nov 2024) | Scaled Agile, Inc. | Nov 2023 |
| Verified International Academic Qualifications | World Education Services | May 2023 |
| New Models of Business in Society | UVA Darden School of Business | May 2020 |
| Machine Learning for Business Professionals | May 2020 | |
| ITIL® Foundation, IT Service Management | AXELOS | Jun 2017 |
I am interested in the seam between systems that are designed and systems that are lived with.
Public services are the clearest example. A tribunal platform is architecture on one side of the screen and someone’s housing dispute on the other. That gap is where most digital government work quietly goes wrong, and it is the part I find genuinely worth doing well.
Outside the mandate work, I build the tooling that reduces the cognitive load of architecture itself: knowledge systems, playbooks, and AI agents that let a small practice behave like a large one. Two of those are open source: the Enterprise AI Framework and AEGIS. Some of that thinking ends up in writing.
If you are standing up an enterprise or portfolio architecture practice, modernizing a regulated platform, or trying to make AI adoption survive an audit, that is the conversation I am most useful in.