Our Technology Stack
Every technology listed here has been deployed on a live, production system for a real client. Not a tutorial project, not a proof of concept.
Book a Free 30-Min CallMicrosoft .NET Platform
We've built government HR systems, education platforms, and business portals with this stack. It handles 150,000+ concurrent users when built properly.
.NET 10 (current LTS)
New builds target .NET 10, the long-term support release Microsoft shipped in November 2025. Production systems we maintain have been upgraded through .NET Framework, Core 3.1, .NET 6, and .NET 8, and we plan the move to 10 the same way. Each upgrade is planned, not just a project file change.
ASP.NET Core and Web API
Primary backend framework across all government and enterprise engagements. REST API design, JWT authentication, claims-based access, and Essential Eight-aligned security practices.
Blazor Server and Blazor WASM
Used for government portals where business logic and identity management belong server-side. Production deployments with Windows Authentication, MudBlazor UI, and mainframe integration.
Entity Framework Core
Code-first migrations, optimised query patterns, and production performance tuning. We've fixed queries running at 8 seconds down to under 200ms using proper index strategy and EF query analysis.
Frontend
Our Angular work includes upgrading Angular 2 applications to current versions without full rewrites. Next.js is used for SEO-critical sites.
Angular (TypeScript, NgRx, RxJS)
Enterprise portal development with Angular 14+ and strict TypeScript. NT Government budget management and worker screening systems. We've upgraded Angular 2 applications to current versions without full rewrites.
React and Next.js
This website runs on Next.js 15 with the App Router and React Server Components. We use Next.js for SEO-critical sites where sub-second LCP scores matter. React for dynamic client-side interfaces.
TypeScript across all frontend work
No JavaScript without types on any project we own. TypeScript catches the category of bugs that cost the most to fix in production, specifically the ones that only show up when real data hits the system.
WCAG 2.1 AA accessibility compliance
Built in from the start on all government work. Semantic HTML, keyboard navigation, screen reader testing, and colour contrast checks are part of the definition of done, not a post-launch checklist.
Data & Databases
We've diagnosed and fixed queries that were taking 8 seconds down to under 200ms. Database performance is usually where the biggest wins are.
SQL Server (2012 through 2022)
Primary relational database across government and enterprise projects. Query optimisation, index strategy, stored procedure analysis, and SSIS data migration pipelines.
PostgreSQL
Used on commercial projects and WhatsApp automation systems. Supabase for managed PostgreSQL hosting where appropriate. Solid performance on high-volume message processing workloads.
ADABAS mainframe databases
Production integration with ADABAS-based NT Government payroll and HR systems via EntireX middleware. This is specialist work with very few practitioners in Australia. We've run it on live payroll systems without unplanned outages.
Azure SQL and managed databases
Azure SQL Database for cloud-hosted projects with data residency requirements. Elastic pools, automated backups, geo-replication for disaster recovery where required by government contracts.
Cloud & Infrastructure
All our cloud work uses Australian data centres where required. We've migrated government systems to Azure with zero downtime.
Microsoft Azure (App Services, Functions, Storage, DevOps)
The primary cloud platform across all our enterprise work. App Service for web hosting, Azure Functions for event-driven processing, Blob Storage for document management.
Azure DevOps CI/CD pipelines
Automated build, test, and deployment pipelines. We've also worked with Bamboo CI/CD in NT Government environments. Deployment gates, environment promotions, and rollback procedures are documented before go-live.
Australian data sovereignty compliant deployments
All cloud work for government clients uses Australian Azure regions. Data residency is documented in architecture decisions, not assumed. We've had government clients require this explicitly and we've delivered it on multiple contracts.
Automation & Integration
We run n8n on Azure Australia so your automation data never leaves Australian infrastructure. Clients have recovered 15-20 hours per week per person.
n8n self-hosted workflow automation
Self-hosted on Azure Australia so your automation data never leaves Australian infrastructure. We build deterministic workflows that extract structured data, route it for approval, and connect systems that were never designed to talk to each other.
WhatsApp Business API (Meta)
Custom membership and notification systems integrating WhatsApp Business API with PostgreSQL and Stripe. Automated communication workflows handling high daily message volumes for Australian businesses.
REST and SOAP API development
REST API design and implementation is standard practice. SOAP and XML-based services are still present in government environments and we've integrated with them on multiple NT Government projects.
Mainframe integration via EntireX connectors
Modern .NET APIs connected to live ADABAS/Natural mainframe services. The mainframe keeps running. The manual workflow layer on top of it stops. On one NT Government HR system this eliminated over 500 manual processing tasks per month.
Architecture & Security
We don't talk about architecture for its own sake. These patterns exist because systems built without them become expensive to maintain and hard to change.
Clean Architecture
Separation of concerns across presentation, application, domain, and infrastructure layers. Systems built this way are cheaper to change 18 months after delivery because the business logic isn't tangled with database calls and HTTP concerns.
Event-driven architecture patterns
Used on systems where components need to respond to state changes without tight coupling. Reduces the blast radius when one part of the system fails or needs to change independently.
Zero Trust security design
Every request is authenticated and authorised, regardless of network location. Relevant for government systems where the threat model includes internal network compromise, not just external attacks.
Claims-based authentication
Production experience with Windows Authentication, WIF claims transformation, and JWT claims in both government environments and commercial applications. Multi-factor authentication as an architectural requirement, not an afterthought.
ACSC Essential Eight alignment
Application control, patching discipline, macro hardening, and MFA are built into architecture from day one on all NT Government engagements. Not retrofitted at the end of a project.
Lessons from Production
18+ years of production deployments on government systems generates insights you will not find in documentation. Here is what the actual work has taught us.
Eliminating 500+ monthly manual workflows without touching the mainframe
An NT Government HR system was generating over 500 manual processing tasks every month because three different systems did not talk to each other. The payroll data lived in ADABAS on the mainframe. The right answer was not to replace the mainframe. It was to build an API wrapper using EntireX middleware and automate the workflow layer in ASP.NET MVC on top of it. The mainframe kept running. The manual work stopped.
HR processing time dropped by 60%. The mainframe integration has been in production for over two years without an unplanned outage. The lesson: solve the workflow problem, not the infrastructure problem you were not asked to solve.
Why Blazor Server was the right choice for a government identity system
A government employee identification system needed to match records across multiple employment episodes and assign unique identifiers, with a full audit trail. We chose Blazor Server over a JavaScript SPA specifically because the matching logic and audit requirements needed to stay on the server. Windows Authentication and claims-based access were already in the government environment. Blazor Server let us keep all of that server-side without building a separate API layer that would need its own security surface.
The system has been processing employee records for 20,000+ government employees since delivery. Zero unplanned downtime. The decision to use Blazor Server was questioned early. It has not been questioned since.
Production at 150,000+ users teaches you things smaller systems cannot
Working as a contractor on a state government education platform used by over 150,000 international students meant that any broken SSIS package was not a development problem, it was a student enrolment problem. The stack was .NET Core 6, Angular 14, Entity Framework, and Azure Blob Storage, with SSIS handling data migration from external systems. Monitoring those SSIS packages, diagnosing failures under load, and managing change requests without breaking live enrolments was the most rigorous production discipline we have experienced.
That scale of production exposure shapes how we approach testing and release management on every project since, including projects a fraction of that size.
.NET version upgrades are never just a project file change
Across multiple government projects we have upgraded production systems through .NET Framework, .NET Core 3.1, .NET Core 6, and .NET 8. Each jump surfaces breaking changes in Entity Framework behaviour, middleware pipeline configuration, and third-party package compatibility. Mainframe integration endpoints and claims-based authentication need specific regression testing after every runtime upgrade.
The rule we now apply: never upgrade the .NET runtime and a major dependency in the same release. Separate them. When something breaks, you need to know which change caused it.
Stack at a Glance
Want to discuss the right stack for your project?
Book a free 30-minute call. We'll look at what you're building, what you already have, and recommend the technology that actually fits the problem.