Software has never been easier to build. But as AI removes the barriers that once made development a highly specialized profession, a more uncomfortable question is emerging: what happens to the people whose hard-won expertise is suddenly accessible to almost everyone?
A few years ago, turning an idea into an application was a serious commitment. You needed a developer, sometimes an entire team, a reasonable budget, and enough patience to wait weeks or months before seeing something functional.
Today, someone can open Lovable, Base44, or another AI-powered development platform, describe an idea in everyday language, and have a working prototype surprisingly quickly.
For entrepreneurs, this is extraordinary. For someone who has spent years learning how to write software, the experience can be a little more complicated.
I’ve been working in technology for years. Computer science is something I studied, something I genuinely enjoy, and something that has shaped the way I approach problems. I use AI extensively, too. I understand what these tools can do and appreciate how much time they can save.
Still, watching software development change at this pace has made me think about something we rarely discuss when celebrating the latest AI breakthrough. What happens to the value of professional expertise when the work itself becomes dramatically easier to reproduce? And what does that mean for the people who built their careers around it?
When a $5,000 Project Suddenly Looks Like a Monthly Subscription
There is something remarkable about the speed of this transition.
A relatively simple application that might once have required thousands of dollars in development can now appear on someone’s screen after a few prompts. To the person using the AI tool, the conclusion feels obvious: why would I pay a developer thousands for something I can generate myself?
The problem is that a convincing prototype and a production-ready application are not necessarily the same thing.
There are security considerations, database architecture, maintainability, performance, integrations, technical debt, and countless decisions that become important when software moves beyond a demonstration.
But those differences aren’t always visible. And that’s where things get interesting.
The actual value of experienced engineering hasn’t necessarily disappeared. What has changed is how that value is perceived.
If someone can generate an application in fifteen minutes, it becomes much harder to explain why a similar-looking project might require several weeks of professional development.
That perception has consequences. It affects pricing, client expectations, hiring decisions and the way people evaluate developers.
I don’t think this means software engineering is becoming worthless. But I do think certain types of development work are losing the scarcity that once made them commercially valuable. And that is a significant shift for an entire profession.
When technology makes a difficult task look effortless, the expertise behind that task can become almost invisible.
The Other Side of Democratisation
Imagine that tomorrow a new automated system becomes available. You place it in your living room, choose a color, and within minutes it paints your entire house. Perfectly, at a fraction of the traditional cost.
As a homeowner, I’d be delighted. I’d save money, avoid the disruption, and probably wonder why anyone would choose the old process.
Now imagine being a professional painter. Someone who has spent twenty years learning the trade, building a customer base and providing for a family.
The exact same technology that represents freedom and convenience for one person could represent a serious threat to another.
Neither reaction is irrational. And that, to me, captures something essential about the current AI revolution.
For aspiring founders, vibe coding is incredibly empowering. People who previously couldn’t afford development teams can finally experiment with their ideas. Someone with no technical background can build an internal tool, validate a business concept, or create a first version of a product.
I genuinely believe that’s a good thing.
But I also understand why professional developers might look at this transformation with mixed feelings. We’ve spent years learning skills that are suddenly becoming accessible to people who have never studied them.
That doesn’t mean those people shouldn’t have access. It means technological progress has a human dimension that deserves more attention.
An innovation can be good for society without being equally beneficial to everyone affected by it.
That’s not an argument against progress. It’s simply an acknowledgment of reality.
Did Software Engineers Automate Themselves Out of a Job?
I’ve heard versions of this argument before: developers built AI, and now AI is coming for their jobs.
There’s an obvious irony in that observation, but I don’t find it particularly convincing.
Computer science was never supposed to be about protecting the professional exclusivity of programmers. It’s about computation, information, systems and solving problems.
If we develop technology that makes difficult tasks easier, reduces costs and allows more people to create useful things, we are accomplishing something meaningful.
AI itself isn’t a recent invention. Neural networks and machine learning have been studied for decades. What changed was the combination of advances in algorithms, computing power, infrastructure, data and investment that made today’s capabilities possible.
We shouldn’t forget that the tools we now take for granted are the result of decades of scientific and engineering progress. It would be strange to celebrate the purpose of computer science while rejecting one of its most consequential achievements.
But accepting progress doesn’t mean pretending that its consequences are painless.
I can appreciate what AI makes possible while also questioning what it means for my profession. Those two positions can coexist.
Building the Product Was Never the Whole Challenge
There’s another issue worth discussing, particularly with the enormous enthusiasm surrounding AI-generated applications.
Building software and building a business are two very different things.
An entrepreneur can now create a functional MVP over a weekend. That’s impressive. But what happens on Monday?
Who is going to use it? How will the founder reach potential customers? Why would anyone pay for it? How will it compete with alternatives? And what happens when the initial prototype needs to become a reliable product?
None of these questions disappears because development becomes faster. In fact, some become more important.
For a long time, technical founders have been tempted to focus on the engineering itself: the architecture, the frameworks, the elegance of the code, the performance.
These things matter, of course. But excellent engineering alone doesn’t guarantee commercial success. You can spend months creating beautifully written software that nobody wants. And you can build a relatively basic product that solves a genuine problem and gradually develops into a successful business.
Understanding the customer, distribution, marketing, positioning and the economics of a product has always been critical. AI is simply making that reality harder to ignore.
When building becomes easier, knowing what to build, who to build it for and how to bring it to market becomes an even more important source of differentiation.
This doesn’t mean everyone needs to become a businessperson. It means that technical execution, by itself, may no longer be enough to establish a sustainable competitive advantage in many areas of software development.
So What Happens to Developers?
I don’t believe developers are about to disappear. I do believe that our understanding of what makes someone a valuable developer is changing.
We’ll continue to need engineers with deep technical expertise: people who understand complex systems, algorithms, infrastructure, security, architecture and the technologies required to build increasingly sophisticated software.
The AI industry itself depends on that expertise.
But for developers whose work primarily involves implementing relatively standard applications, the pressure is likely to increase. And I don’t think it helps anyone to ignore that possibility.
The next generation of developers may need to become more involved in product decisions, business requirements and the broader context surrounding the software they create. Not necessarily founders. Not necessarily entrepreneurs. But professionals who understand the problem before choosing the solution.
That is also where technical experience continues to matter. Someone who understands software engineering can use AI very differently from someone who has only learned to describe what they want.
An experienced developer can question the proposed architecture, recognize technical risks, evaluate trade-offs and identify problems that may not be visible in a working prototype. The difference isn’t simply who writes a better prompt. It’s who understands the consequences of the decisions being made.
Computer science education will also need to adapt, maintaining strong technical foundations while incorporating more practical exposure to AI, product development and the commercial realities of technology.
The Human Side of an Accelerating Industry
What concerns me most isn’t that technology keeps improving. That’s exactly what technology should do.
It’s how quickly the economic value of certain skills can change.
People spend years studying, practicing and developing expertise. They build careers around that expertise. They make financial commitments, raise families and develop a professional identity.
Then, sometimes within a remarkably short period, the market starts questioning the value of the work they perform.
We often discuss these transformations through productivity statistics and business opportunities. We talk about efficiency, lower development costs, faster deployment and increased output. All valid considerations.
But behind those numbers are people trying to understand where they fit into a changing economy.
An AI model can become significantly more capable in a matter of months. A person cannot necessarily rebuild an entire career in the same amount of time.
That’s why I believe the conversation around vibe coding needs to extend beyond demonstrations of what the latest tools can generate. It’s also a conversation about education, professional identity, economic opportunity and the speed at which society can adapt to technological change.
And it doesn’t stop with programming. Similar questions are emerging in design, marketing, writing, analysis and many other professional fields.
We may be looking at a broader transformation in how society values specialized knowledge.
Learning to Build in a World Where Everyone Can Build
I remain optimistic about technology. I enjoy working with AI, experimenting with new tools and discovering ways to solve problems that previously required much more time.
There’s something genuinely exciting about the fact that a person with a good idea no longer needs substantial resources to start building. We should celebrate that.
At the same time, I don’t think we should dismiss the concerns of professionals whose work is being transformed. Recognising those concerns isn’t resistance to innovation. It’s part of understanding its consequences.
Perhaps the most difficult thing about this moment is accepting that two seemingly contradictory ideas can both be true.
AI is democratizing software development. And AI is challenging the economic value of certain software development skills.
The opportunity is real. So is the disruption.
I don’t know exactly what the software industry will look like ten years from now. I suspect some of our current assumptions about technical careers will seem outdated, while entirely new opportunities will have emerged.
But I do think one thing is becoming increasingly clear.
The ability to build software is no longer as rare as it used to be. The ability to understand what should be built, and why it matters, remains much harder to replicate.
Maybe that is where the definition of a developer begins to change. Not someone whose entire professional identity depends on writing code, but someone who understands technology deeply enough to use it in solving meaningful problems.
Software is becoming easier to create. Understanding people, making good decisions and building something that genuinely matters is still difficult.
And perhaps that’s where the next chapter of software development really begins.