When low-code is not the right tool, we write real code: tested, hosted in your Azure subscription, and built so someone else can maintain it.
Low-code is the right answer for many problems, and we use it often. It stops being the right answer when the logic is too involved to test inside a flow, when two systems need to stay in step through failures and restarts, or when an application has outgrown the platform it started on.
In those cases we write code. Business rules go into tested libraries, not formulas buried in a spreadsheet or a canvas app. Integrations retry, match records and reconcile, and they log what happened. Users get a simple screen to review the result and override it when they disagree.
We also take over software that another developer built and no longer supports, and we run three products of our own on the same stack we build for clients: DevFlow for consultancies, StadiumCue for school sports broadcasts and Meal Planner for groups. That stack is .NET and C#, Python, TypeScript and React, and Azure, with custom connectors that bring the result into Power Platform when that is where your team works.
Business rules written as tested code and checked against your existing figures, with a review app where people approve or override the output.
Scheduled and event-driven syncs between business systems that retry failures, match records and reconcile the two sides.
Web APIs for your own data and custom connectors that expose them to Power Apps and Power Automate.
TypeScript and React front ends on Azure Static Web Apps, with .NET or Python back ends and sign-in through Microsoft Entra External ID.
We review code another developer wrote, close security gaps, fix defects and keep adding features without a rewrite.
Azure App Service, Functions, SQL and Key Vault set up in your subscription, with monitoring and alerts on the jobs that matter.
We confirm that code is warranted and that low-code, an existing product or a smaller change would not do the job.
Rules and integrations are written with tests, and calculations are compared against the figures you already trust.
People see the system results in a simple app where they can approve, correct or override before anything is final.
The system is deployed to Azure with logging and alerts, and we support it afterward or hand it to your team with documentation.
Clients are described by type. We do not publish client names.
The problem. Ordering decisions lived in a spreadsheet that nobody could maintain.
What we built. The calculation rebuilt in tested code, confirmed to reproduce the original workbook figures exactly, behind an app where managers review and override recommendations.
The problem. A platform started by another developer had security gaps and defects, and the business needed new features.
What we built. We took it over, closed the security gaps, fixed the defects and added new modules without a rewrite.
The problem. Referral records in the clinical system had to reach the CRM without manual re-entry.
What we built. A scheduled job that sends each record across, logs every success and failure, and continues past individual errors.
What we build with: .NET and C# · Python · TypeScript and React · Azure App Service · Azure Functions · Azure Static Web Apps · Azure SQL · Azure Key Vault · Microsoft Entra External ID · Power Platform custom connectors · n8n
When the logic is complex enough to need real tests, when an integration has to survive failures and resume cleanly, or when you need an API or a user experience the platform cannot give you. If low-code can do it well, we will say so.
You do. It lives in your repository and runs in your Azure subscription, and we document it so your team or another firm can maintain it.
Yes. We start by reviewing the code, hosting and security, then fix what is urgent and add features in small steps rather than proposing a rewrite.
We deploy to Azure with logging and alerts on the jobs that matter. After launch we can keep supporting the system, or hand it over with documentation.
AI coding assistants are used with tests written first and independent review before anything ships. The tests and the review, not the assistant, decide what goes live.
Free 30-minute call. No sales pitch. We'll tell you honestly what we would do and whether we are the right fit.
