The Truth About The Impact of AI on Software Development
During the World Government Summit in Dubai earlier this year, Jensen Huang (Nvidia CEO) said something intriguing; “Over the course of the last 10-15 years, almost everybody who sits on a stage like this would tell that it is vital that your children learn computer science, everybody should learn how to program. And in fact, it is almost exactly the opposite.” Did that cause a stir? To some extent, yes.
Katie Mikova
8min read
During the World Government Summit in Dubai earlier this year, Jensen Huang (Nvidia CEO) said something intriguing; “Over the course of the last 10-15 years, almost everybody who sits on a stage like this would tell that it is vital that your children learn computer science, everybody should learn how to program. And in fact, it is almost exactly the opposite.” Did that cause a stir? To some extent, yes.
But is he right? Not entirely because the reality is more complex and the impact of AI on software development jobs is still being observed and debated.
Generative AI, Low Code, & The New Era of Development
There are now autonomous generative AI systems, like Qodo (previously CodiumAI), offering an AI assistant for generating code, writing unit tests, and creating documentation. Then, we have low-code tools like App Builder that bring AI features (App Builder AI), UI components, code generation, and more to streamline development cycles.
Read more in App Builder AI: How Does It Streamline App Development
Although AI-powered code generation tools have already modified how developers write code, there is a long way to go before we observe a complete takeover of software development jobs. Even though software development dynamics have changed significantly, it’s hardly an AI revolution for now.
The truth is rather than eliminating the role of developers, AI serves to augment it in a way that:
- Affects less experienced programmers more than expert developers
- Requires teams to carry out architectural direction and quality checks
- Enables junior developers to handle mundane and repetitive tasks
- Speeds up programming – writing more code faster
- Releases the next productivity boom but demands upskilling and adaptation
- Accrues business value, agility, and growth
So, computer programming is still expected to remain a human-centric pursuit for the most part. And if there’s one thing I agree with what Jensen Huang said, it’s this: “Everybody in the world is now a programmer. This is the miracle of Artificial Intelligence. You now have a computer to do what you tell it to do.” Yet, human capacity and capabilities remain at the core of software development.
As McKinsey points out in The Economic Potential of Generative AI report: “The era of generative AI is just beginning. Excitement over this technology is palpable, and early pilots are compelling. But a full realization of the technology’s benefits will take time, and leaders in business and society still have considerable challenges to address. These include managing the risks inherent in generative AI, determining what new skills and capabilities the workforce will need, and rethinking core business processes such as retraining and developing new skills.”
The Changing Role of Software Developers
Some initially feared that generative AI and low-code platforms might replace traditional programmers and render coding skills obsolete, which is a misconception. On the contrary, low-code development and AI aim to streamline the process by automating repetitive tasks so developers can focus more on core business logic and feature development, reducing the time spent on repetitive technical tasks and collaboration challenges.
McKinsey indicates that; “Generative AI has the potential to change the anatomy of work, augmenting the capabilities of individual workers by automating some of their activities. Generative AI and other technologies have the potential to automate work activities that absorb 60% to 70% of employees’ time today.”
Consequently, with the impact of AI on software development, there is a shift that fosters a mix of technical and non-technical skills. Developers learn to bridge the gap between stakeholders, translate requirements into modular components, and identify where and how automation can be applied most efficiently.
Moreover, with low-code platforms and AI, more opportunities are now presented to a broader pool of experts, even non-technical people, as these tools eliminate the need for hand-coding. We see the rise of citizen developers who have become increasingly involved in app development. C-level executives and stakeholders learn how to use low-code tools as instruments that bring together fusion teams and boost development productivity.
The role of developers, on the other hand, now includes the oversight of low-code projects, code inspection (as low-code tools like App Builder generate code for popular frameworks like Angular, Blazor, Web Components, React), guiding best practices, and more.
Is Programmers’ Job Security at Risk Then?
Not yet, because there are certain AI limitations. AI advancements and low-code development don’t directly threaten job security but instead, shift the type of work developers do. While these platforms automate repetitive coding tasks, developers are still essential for complex system design, customizations, and high-level integrations that low-code solutions can’t handle independently.
The reduced need for manual coding frees developers to focus on higher-value tasks like architecture, problem-solving, and more strategic roles within the project lifecycle. As companies increasingly adopt AI and low code to accelerate delivery, developers who understand how to optimize and extend these platforms remain vital. Their expertise evolves into roles where they guide and manage AI-driven workflows, thus enhancing their job stability.
How What Is The Impact Development Workflows?
Low code and AI’s impact on software development jobs work like this – they streamline workflows by enabling faster prototyping, reducing the initial coding workload, and simplifying the feedback cycle through visual, often WYSIWYG interfaces that stakeholders can review early on. It changes the workflow by allowing more collaborative and iterative development; non-developers can contribute, which speeds up requirement gathering and early-stage revisions.
However, these automation platforms also require oversight to ensure that generated components align with broader architecture and performance standards. Overall, it accelerates project timelines but also demands a balance, as developers must sometimes intervene for customizations and optimization.
What AI Can (and Can’t) Do in Software Development
Currently, coding through AI is not yet optimal. There are still context limitations within which AI and low code operate. These tools are best suited for generating small, simple, and independent code pieces, which aligns with programming best practices—writing reusable, simple, and modular code.
| Aspect | Generative AI | Low Code |
|---|---|---|
| Coding Approach | Probabilistic Model When using NLP models | Deterministic Model |
| Code Generation Consistency | Results vary with each prompt, even when the same question is asked multiple times | Produces the same output for the same task, ensuring consistency |
| Scope of Use | Best suited for generating small, simple, and independent code snippets | Suitable for building automated, consistent, end-to-end solutions |
| Context Awareness | Lacks understanding of how code snippets integrate into a larger system | Operates with a comprehensive understanding of the entire system, allowing it to generate cohesive code solutions |
| Alignment with Best Practices | Supports writing modular and reusable code but lacks holistic project awareness | Ensures a consistent approach to system development, adhering to automated programming practices |
| Limitations | Limited by context, requiring human intervention for complex integrations and customization | Cannot provide fully customizable solutions but excels at systematized, structured development |
Zdravko Kolev, Product Development Manager here at Infragistics, points out that; “Senior developers use AI more effectively—they ask better questions and, in return, receive better, more specific answers. Junior developers who provide less context tend to receive more general responses. Using Gen AI or low-code solutions is highly beneficial for repetitive tasks as we obtain general, existing functionalities that make developers more productive. This enables faster and higher-quality output in less time.”
While the impact of AI on software development is less severe, low code (though extremely powerful) still has certain limitations when delivering fully customizable solutions. Developers, then, aren’t at risk of becoming obsolete.
What Happens With Quality and Security In Programming?
The AI’s impact on jobs is that it promotes faster software rollouts. This means that AI increases the amount of written software, but it does not necessarily increase the quality. The result is often difficult to maintain spaghetti code without the quality and structure required by a high-performance mission-critical app.
Over time, fewer people will write code manually. Especially for front-end solutions, low code provides significant value in a short time. Developers will increasingly focus on knowing how to ask the right questions to generate higher-quality code. Eventually, the impact of AI on software development will urge developers to evolve into skilled, prompt engineers, knowing precisely how to guide AI or low-code tools to achieve the desired result.
Nevertheless, companies will still need people who can understand the code and programming concepts, review it, maintain it, and ensure the app meets best practices, architecture principles, and coding standards.
