Best Practices

In every mature .NET landscape, legacy projects represent both heritage and hazard. They once powered entire business models — now they silently consume time, budget, and attention. The decision to retire or modernize them isn’t about technology fashion. It’s about sustaining the organization’s capacity for value creation.

In a previous article, we laid it out – unfiltered: Copilot turns junior devs into syntax secretaries.

Not because it’s evil. But because it removes friction before understanding.

It gives you working code before you know what working even means. It creates the illusion of progress, while slowly eroding the very skills that define a software engineer: reasoning, decision-making, and technical ownership.

The hype around GitHub Copilot (or any other AI code assistant) is deafening. AI-assisted coding. Effortless automation. 10x productivity.

But here’s the harsh truth: Copilot isn’t empowering junior developers – it’s deskilling them.

In the world of software development, there’s a recurring tension between discipline and improvisation. Somewhere along that spectrum lies a phenomenon increasingly referred to as Vibe Coding. The term evokes a style of development where engineers follow intuition and momentum rather than formal plans, processes, or design patterns.

It’s fast, fluid, and occasionally brilliant. But is it sustainable in a .NET-based enterprise context?

When we activated static code analysis for the first time in one of my last projects, the overwhelming number of warnings exceeded expectations and highlighted gaps in the code. Without making any changes, the project already had a significant number of warnings. After activating additional analyzers and updating some configurations, this number temporarily increased dramatically.

The high number of warnings was initially daunting, but we saw it as an opportunity to significantly improve our code quality. At first glance, it seemed easier to suppress or ignore these warnings. But as I often remind my team, “The code you create is a valuable legacy, so it’s important to build it carefully.” Ignoring warnings today creates obstacles for future developers—and that could very well include you six months down the line.

This experience reinforced the importance of managing warnings and errors systematically. Let me share some of the lessons we learned, the strategies we used to tame those 60,000 warnings, and how you can apply these techniques to your own projects.

In software development, there’s a silent debt that accrues interest over time, often hidden beneath layers of code and decisions made in haste or ignorance. This debt is aptly termed technical debt. Much like the german proverb, “Wer den Pfennig nicht ehrt, ist den Taler nicht wert”, (or the english equivalent, “A penny saved is a penny earned”) technical debt reminds us that small oversights or compromises in the present can snowball into significant challenges down the road. This article critically examines the parallels between financial principles and technical debt, emphasizing the importance of addressing both direct and indirect debt while understanding its distinction from external risks such as hacking or abuse.

In software development, dependencies are inevitable - any project worth its salt relies on various libraries, frameworks, or packages. However, as I found in my own work, managing these dependencies can be an onerous task. Constant updates, new vulnerabilities, and endless manual approvals were draining my time and focus. What if, I thought, these processes could be automated? This thought led to the creation of dependamerge, a GitHub Action designed to free developers from the drudgery of manual dependency maintenance and let us get back to what we do best: building great software.

As developers, we’re often tasked with maintaining and modernizing legacy codebases that were written long before some of the best practices of today—such as nullability annotations—were available. While modern C# now supports nullable reference types, enabling us to avoid the dreaded NullReferenceException, introducing this feature to existing, large codebases can be a challenge.

In this article, I’ll share my step-by-step approach for introducing nullability into a legacy .NET and C# project. You’ll learn how to apply nullability in a controlled, incremental manner using project-level settings, scoped annotations, and file/method-level directives, all while maintaining the integrity of your legacy codebase. After all, modernizing your code doesn’t have to be an all-or-nothing endeavor—gradual change is key to a successful transition. Let’s get started!

In the ever-evolving world of .NET development, managing project configurations effectively is crucial for maintaining a clean and efficient build process. One of the less frequently discussed but highly useful properties is BuildingInsideVisualStudio. This property, when correctly utilized, can streamline your build process and ensure that your project is configured properly depending on the build environment. In this article, we’ll explore the BuildingInsideVisualStudio property with concrete examples and discuss best practices for using it effectively.