Is AI Going to Replace Programmers?

The Rise of AI in Software Development

AI is turning out to be the keyword of this decade—and for good reason. I earned my degree in Computer Science before the age of Large Language Models (LLMs), and now I use them daily. What does software as a career look like going forward?

Software as a career is not going away any time soon. But it is changing, now.

What Is AI? 🤔

For those unfamiliar with the details or overwhelmed by the buzz, the key thing to understand about AI in today's discussions is this: the AI applications people refer to in the discussion about coding refers to LLMs (Large Language Models). ChatGPT, Gemini, etc. At the core, LLMs are advanced prediction models—they predict the next most likely word or sequence based on patterns in vast amounts of data. The practical implication? Free-form critical thinking is not their primary function. If you're interested in a deeper explanation, I recommend this article from Google: Introduction to LLMs.

What Is Coding, at Its Core? 🎯

Strip away the syntax, frameworks, and tools, and coding boils down to one thing: problem-solving. Writing software is about breaking down challenges, structuring solutions, and translating those solutions into instructions that a computer can execute.

Career Trends? 💼

No crystal ball, but here is my assessment of current trends :

1. The Value of Excellent Thinkers is Growing

If we put two and two together, it's clear that programmer productivity is increasing, and will continue to do so, as our toolkits improve. So naturally the productivity expectation of programmers in the market will increase as well. However, the parts of coding that AI is replacing are not the critical thinking aspects.

2. The Age of AI-Assisted Coding

Anyone who has spent a meaningful amount of time using LLM coding tools will know that they are good, and are here to stay. I remember when ChatGPT made the news in 2022. I challenged myself to use only ChatGPT to help create a Python program. At the time, it was much slower than if I had used Google instead. We are long past those days. These LLMs are already more than acceptable, and are continuing to push boundaries.

3. T-Shaped Programmers Shine

In software, value is driven by problem solvers, not just those who know how to use a specific toolkit. The industry moves too fast for surface-level expertise to be enough—AI has only accelerated this shift. This is why "T-shaped" programmers stand out: they can adapt to different environments while developing deep expertise in a specific area.

An HVAC Analogy

To bring these idea to life, let’s use an analogy. Imagine your HVAC system breaks down, and YouTube tutorials aren’t enough to fix it. The issue is complex—so what qualities would you look for when hiring a repair professional?

1. You’d want someone who can diagnose the problem and determine the right solution.
2. You’d want them to have the right tools and professional equipment.
3. You’d prefer an HVAC specialist over a general handyman.

Similarly, in software, success comes from problem-solving ability, competence with the right tools, and specialized expertise.

Questions?

Contact us below! We’re happy to discuss all things software. Follow us on LinkedIn for updates on new blog posts.