National Coding Week: Software development doesn’t look the same anymore

Coding is not just about mastering programming languages anymore; it’s about working with intelligent systems that can design, debug, and even deploy applications.

This year, National Coding Week arrives at a pivotal moment. We hear from tech industry experts reflecting on how AI has transformed a developer’s role, what risks must be addressed, and how the next generation will be prepared for a future where humans and machines code side by side.

Vibe-coding is the new era

Matt Moore, CTO and co-founder of Chainguard (software supply chain security firm), says the pace has taken many by surprise. “The way we create software is evolving far more quickly than most of us expected. Gartner predicts that 40% of new business software will be AI-assisted in the next three years, and that's a conservative estimate.”

He mentions that vibe-coding isn’t just a passing trend, but a fundamental change in how code is developed, tested, and deployed.

This new reality is already playing out in the tools developers are using. Dr. Greg Benson, Chief Scientist at IPaaS experts SnapLogic and Professor of Computer Science at the University of San Francisco, points out that we have gone from basic coding assistants to more advanced coding agents.

“Coding assistants, like the early version of GitHub Copilot, provided code completion inside an IDE (Integrating Development Environment) that anticipates what code might come next as a developer is writing code. However, in the last year, we have seen the rise of coding agents. These can autonomously carry out complex coding activities like creating applications from scratch or modifying existing code. In addition, the coding agents can help develop application requirements, implementation plans, and also help debug problems in code.”

Speed vs Security: The tension for modern developers

According to Dr. Benson, developers are leaning on autonomous systems to get work done remotely and in parallel, with speed and convenience driving this adoption.

“We are beginning to see mobile-based experiences in which you can work with your coding agent remotely from your phone. So you can effectively plan, develop, run, and debug through text or voice at anytime from any location. In addition, it is possible to coordinate multiple coding agents simultaneously and try many different plans in parallel and then the agent can help pick the best result. LLM-based Generative AI continues to revolutionise software development as we know it.”

But this drive toward greater speed and autonomy comes with challenges. Scott McKinnon, UK&I CSO at cybersecurity firm Palo Alto Networks, says developers today are under enormous pressure to move quickly. 

“The unsanctioned use of AI tools by developers is becoming a serious issue as application development continues to evolve at a rapid pace. Building modern, cloud-native applications isn’t just about writing code anymore - it’s about a delivery model operating in continuous beta mode. The knock-on effect is that developers are under intense pressure to be fast and reduce time to market. With this in mind, it’s not surprising that many developers are using AI tools in an effort to increase efficiency and deliver on these challenging expectations.”

Embedding oversight and accountability

McKinnon explains that, in many cases, developers are bypassing guardrails to meet deadlines. “Many developers are seeking ways around the organisation’s AI guardrails. This too often means spinning up open-source LLMs outside of approved platforms, using AI to generate code without oversight, or skipping data governance policies to speed up implementation. While the intent behind this kind of efficiency could be considered admirable, the outcome can be costly. Intellectual property gets exposed, compliance slips, and security gaps are created.”

Moore adds caution, saying that without strong safeguards, this reliance on AI could introduce vulnerabilities. “AI functions like a tireless, no-ego junior engineer - highly skilled, but requires constant oversight and guardrails. Without those guardrails and oversight, we risk introducing vulnerabilities by omission.”

He stresses that as AI takes on more responsibility, the developer’s role must adapt. “As we mark National Coding Week, it’s clear that the future of coding goes beyond learning new languages or frameworks, it’s about mastering how to collaborate with AI responsibly. The focus must shift to how we approach quality, review, and security; ensuring every line of code, whether human-written or AI-generated, is trusted and secure.”

“If we can make the safe path the easiest one, our software will not only be faster and smarter, but also secure by default. AI might be rewriting the rules of coding, but we must be upholding transparency and accountability at every step.”

Preparing the next generation of coders

While today’s developers navigate this balance, Simon Black, Director of Evangelist Programmes at low-code app development platform Mendix, urges us to look at the next wave of coders, the children growing up surrounded by AI tools.

“National Coding Week is not only the perfect opportunity to reflect on significant changes in the coding world in the past year, but also anticipate what it might look like in the future. I believe the future of coding is set to be less about typing lines of code and more about conversation and collaboration. The shift will move from the textual to the visual, where developers will interact with AI as a team member rather than just a tool. We will see a greater use of low-code and no-code solutions as AI handles the heavy lifting, allowing humans to step into a more managerial or product-focused role.”

He believes teaching children about intent-based development early will be crucial.

“With this in mind, we should be teaching the future generations about intent-based development and how to interact with AI. Introducing them to AI early on will help them bring their imagination to life. My eldest child is five years old, but already talks to AI and asks it to make pictures of things he has in his head or ask it questions. Just like my child does, it’s only a matter of time before we see children building applications visually and interacting with AI to do so. This is why low-code as a way of abstracting and explaining code is so important. Children are visual learners, so AI and low-code is going to be crucial to their development. They are the future of coding, so the responsibility is on us to support them as they pave the way.”

Dr. Benson concludes with a message of reassurance for students and future coders alike. “I tell my students that AI won't eliminate the need for software developers, but it changes how software developers do their work. It is more important than ever for new computer science students to become experts in the field in order to work with AI coding agents as colleagues to collaborate to build next generation software.”