Software-Dev

Why are some developers still not embracing AI coding tools? Adopting new tools is inherently risky. This is true of many new tools, but it is especially true of AI coding tools, which require a fundamentally different way of working. So why do some teams embrace these tools while others remain skeptical? In my experience, it comes down to two things: time to think and the ability to fail. Time to Think Learning something new requires time to explore, experiment, and learn. People need. the space to explore without a looming deadline. With deadlines, people fall back on what they know is reliable and will work. ...

2026 is going to be an exciting year. We saw just the tip of the iceberg on what AI coding tools can do in 2025, and now the floodgates are open. AI has fundamentally changed how fast software can be built. This means smaller teams can ship higher-quality software and quickly. We are going to see an explosion in new software products hitting the market in all categories. With increased competition, what will rise above the noise? Quality will be King. When everyone can build fast, customers will choose the best products. ...

Is it wise to start a Computer Science degree in 2026? Is AI going to take all of the software development jobs? These are especially relevant questions for me as my son graduates from high school this year. This is the advice I gave him. I believe it is a fine choice to take Computer Science in the year 2026, provided the following two things are true: You love it - The days of this degree guaranteeing a good-paying job are over. If you are just in it for the money, take up a trade which pays well and is in high demand. Competition for jobs, especially at the entry level, will remain high, so this is only a place for people who love it. You have the mind for it - If you don’t have strong critical thinking and problem-solving skills, you are better off looking elsewhere. The mediocre or below-average software developer will be the first to be displaced by AI. I graduated with my Computer Science degree at a funny time. The year was 2003, and we had just navigated through the dot-com bust. Similar to the market today, it was very challenging to find work. For the next few years, the common wisdom was “Don’t take Computer Science”. High School guidance counsellors were advising students to stay away. I remember visiting my University and was shocked to see how much the CS school had shrunk. Just a few years later, in 2008, I was having a terrible time trying to hire people. There were simply too few people with the skills required. ...

Recently, I wrote about the value of senior developers in an AI world. What are the most important skills a software developer should focus on in this world? Here are my top 3… or perhaps top 7 because the last one has 5 parts. System Architecture and Design Context is king when it comes to designing systems. You may have planned the most elegant and “correct” architecture, leveraging all of the greatest new advancements and patterns used in Big Tech, but you are only wasting your time if you are building an MVP at a 2-person startup. AI still struggles to understand when to apply what architectures. It is important to understand what you need today and also the patterns you want to apply in the future. ...

Will AI replace senior developers? This is a common question people are asking today as AI coding tools become increasingly sophisticated. Instead, the aspects that make a senior developer a great developer remain unchanged in the world of AI. It has been said by many that AI raises the floor and not the ceiling. What this means is that the best developers will become even more differentiated from the pack. While AI can help anyone write basic code, it takes deep experience to architect systems, make strategic technical decisions, and navigate complex trade-offs. ...

Startups are under a lot of pressure to deliver fast. This often necessitates the accumulation of technical debt. Many articles have been written on why it is essential to pay off technical debt. Fewer speak to circumstances for keeping debt or evaluate what areas of tech debt should be prioritized. In my experience, software developers often overvalue the benefits and impact of fixing tech debt issues. A developer might take a full day developing a script to save 30 minutes of a one-time repetitive and menial task. Another mistake developers often make is justifying tech debt based on an ideal outcome that may not be realistic. In an attempt to simplify the code or architecture, a developer may find that the “fixed” version is only marginally better and does not yield the full benefits. ...