Software Engineering Principles and Practices
Software engineering is the practice that turns code into systems people can rely on. The distinction matters because typing code has never been easier — AI assistants produce syntactically valid output in seconds — yet the rate at which production systems fail, leak data, or quietly accumulate maintenance debt has not improved. The discipline lives in the gap between code that compiles and code that survives contact with reality.
The articles in this collection treat software engineering as a profession, not a productivity exercise. The recurring theme is the feedback loop: write code, watch it fail, understand why, refine thinking. That loop cannot be automated because closing it requires learning from production failures and applying that knowledge to prevent the next one. Prompt engineering optimizes for speed; engineering optimizes for survival under conditions the original author did not anticipate.
Topics range from defensive programming with ArgumentNullException.ThrowIfNull and guard-clause patterns, through structured logging that does not lie about what happened, to multi-framework targeting decisions that look harmless and quietly break the build on the third project that consumes the library. Clean Code is treated as a starting point rather than a creed — most teams that quote SOLID rarely apply it consistently, and the articles examine what actually works in production versus what looks defensible in code review.
A second cluster of articles addresses the economic reality. Technical debt compounds like financial debt, and small shortcuts become the dominant cost driver three years in. Retiring legacy projects, illuminating debt with analyzers, and recognizing when a refactor is cheaper than another feature release are covered with the trade-offs named explicitly.
The voice across these articles is opinionated and grounded in specific failures. Generic advice rarely changes behavior. Specific failure modes, named clearly, do.

Six Ways ILogger Silently Fails in Production

The Code You Write Today Is Someone's Problem Tomorrow

Stop Pretending TimeProvider Doesn't Exist

AI Code Review Is a Sycophant
