Software developer and agile consultant at Adaptavist.com Jon Kern staked his entire career on practicing and promoting flexibility and speed. As one of 17 independent software practitioners who famously co-authored the Manifesto for Agile Software Development in 2001, he codified a set of values for building software, ones centered on people, collaboration and delivering working results. A quarter century later, The Forecast reconnected with Kern to discover why the Manifesto has never been more relevant.
"My phrase this year is: It's more relevant than ever,” he told The Forecast in April 2026, during the rapid rise of enterprise AI and the use of AI to build and power new applications.
“You need to understand agility more than ever because baked into that is the fact that you need to be technically competent, as vibe coding with AI really allows so many of the values and the principles to come to life."
Harking back to Kern’s first interview with The Forecast in 2020, it’s clear that the rigor it took to succeed before AI is only tightening. The drive to constantly do better, be leaner and more efficient and effective as a developer.
“It’s not a bunch of separate little things,” he said, prior to the AI boom. “You’re part of a system. You’re part of a whole.”
While artificial intelligence is expanding the possibilities and increasing speed, it requires software developers to think holistically.
Kern argues that AI in software development isn’t just about better tools. More importantly, it’s about collapsing the distance between ideas and reality. He’s witnessed the evolution of scalable IT infrastructure, cloud computing services, and cloud native capabilities, shrinking that distance. And now, with the rise of AI-powered tools and processes, that gap is closing, Kern explained in his recent interview with The Forecast.
For developers who want to work faster without sacrificing performance, “there’s no excuse anymore,” he said.
From Overnight Builds to Instant Feedback
Kern was thinking about “agility” long before it became a buzzword. While designing a Manufacturing Execution System for IBM in the mid-1990s, his goal was to finish coding, hit “build,” and come back the next morning to something that worked. “It should be so easy that even I can do it.”
Back then, that required considerable patience and lots of coordination with system administrators. For most developers, infrastructure was complex, fragile, and difficult to access.
Today, reliable builds are more readily available. Cloud computing services and software-defined IT platforms, such as hyperconverged infrastructure systems, have abstracted much of the underlying complexity. As a result, IT teams can now spin up a sandbox, staging, and live production environments to build, test, and deploy applications that scale globally — without ever touching physical hardware.
Kern thinks this shift has redefined how organizations operate.
“Agility is a means to an end,” he told The Forecast, referring to the core tenet of the software manifesto he coauthored 25 years ago. “That end is getting value out in front of my customers’ hands faster and more effectively, and with higher quality.”
He said easier access to IT resources increased the speed and quality of software development. Things now happen in minutes rather than days or weeks because it takes fewer steps and less time to secure the necessary resources.
The AI Inflection Point
Now, Kern said AI is accelerating everything that developers can do to leverage available IT capabilities. For example, he’s been experimenting with tools like Replit — using AI not just to assist with coding, but to fundamentally reimagine how to explore and validate ideas. This brings to bear advances not just in speed, but also in creative freedom.
He saw that developers were once reluctant to explore multiple approaches due to time constraints. Now, AI is making experimentation effortless. Kern described how he can generate multiple working versions of a feature, present options to users, and iterate quickly — all in real time.
“The calculus is radically different,” Kern told The Forecast. “AI empowers developers to be much more nimble and less worried about the sunk cost.”
That unlocks a more fluid, collaborative style of working in alignment with the original principles of agile software development.
“We can stop talking about paper designs,” Kern continued. “Now I can just write code, and you can try it, even getting Replit to create multiple UX concepts we can discuss and choose from is now an easy thing to do.”
AI in Software Development: Amplifying, Not Replacing
For all its promise, Kern is careful not to overstate AI’s emerging role. It is not a shortcut around experience. In his view, AI doesn’t replace good engineering; instead, it amplifies whatever already exists.
“AI will accentuate what you’ve got going on, including bad stuff — or good stuff, because AI doesn’t care,” Kern said.
That means experienced developers still have a critical role to play. Their human understanding of architecture, testing, and design becomes even more valuable in an AI-driven DevOps workflow.
Kern has seen this firsthand. While working with AI-generated tests, he has found himself correcting flawed assumptions and guiding the system toward better practices. When he introduced behavior-driven development (BDD), “it made up some magic answers,” he said, adding that it took some effort to get the AI to apply the principles correctly. Sometimes, “you have to coerce it.”
Simply put: AI can accelerate outputs, but expert human judgment still matters. It requires strong engineering discipline. It puts more emphasis on best testing practices, architectural principles, and understanding behavior-driven development. These are the guardrails that keep AI-generated output from becoming a liability.
"Trust but verify," Kern said, is the operating principle for this moment.
The Rise of ‘Always-On’ Development
One of the most striking changes Kern has seen is the way in which development has become untethered from traditional environments. Because of innovations like cloud-native capabilities and more readily available data center resources, developers can work from anywhere — even from the back seat of a family car. Kern described testing a mobile feature on his phone while he was in the third row seats during a drive, spotting an issue and fixing it on the spot using an AI-powered development environment.
“There’s a lot more flexibility there,” he said. “For better or worse.”
This always-on capability isn’t about encouraging nonstop work. Rather, Kern’s point is about enabling responsiveness. Because most AI applications are developed with cloud-native architectures and containerized to run across different IT infrastructures, flexibility, portability, and control are built in from the beginning.
AIOps and the New Collaboration Model
The relationship between developers and IT teams is evolving as AI becomes embedded not just in development, but also in operations, Kern said. These groups have traditionally operated in separate silos. Developers built applications; operations teams managed infrastructure.
Now, those boundaries are blurring. With AI in a modernized infrastructure, developers can discover “even more powerful ways that your knowledge can contribute to delivering value,” Kern said.
AI can support and streamline the implementation of best practices. It can automate testing and enforce guardrails. Experienced engineers, in turn, can bring to bear their expertise to create systems that are faster and more reliable.
Familiar Principles for a New Era
As software development enters a new chapter authored increasingly by AI, Kern grounds himself in the ideas that first defined his career, when he co-authored the Agile Manifesto’s foundational principles. With the rise of AI, those values are increasingly relevant, he observed.
“You need to understand agility now more than ever because that AI train is going to roll you over with speed,” he warned.
“And if you don't understand that you can use it to your advantage and be more agile than ever, or if you're ignoring things like the customer or the individuals over process and tools, you can easily steamroll into trouble."
He advised developers to mind the gap.
“It’s the gap in time between taking an action and getting real feedback.”
By focusing on that as the practical goal and tapping the power of a modernized infrastructure, “you can use it to your advantage and be more agile than ever,” Kern said.
Adam Stone is a journalist with more than 20 years of experience covering technology trends in the public and private sectors.
Ken Kaplan contributed to this story. He is Editor in Chief for The Forecast by Nutanix. Find him on X @kenekaplan and LinkedIn.
© 2026 Nutanix, Inc. All rights reserved. For additional information and important legal disclaimers, please go here.
Related Articles
