AI is transforming software development – JetBrains CEO Kirill Skrygan says it’s up to developers to transform with it

There may still be a place for strong developer progression in the age of AI, if workers can adapt to rapid changes

A photo of a young, female software developer presenting code to two colleagues on a conference room screen.
(Image credit: Getty Images)

Software developers needn’t worry about obsolescence in the age of AI, according to JetBrains CEO Kirill Skrygan, but there’s no doubt roles and requirements will change as a result of the technology.

AI has taken software development by storm over the last two years, with developers now able to draw upon an ever-expanding array of powerful tools and solutions that can generate or rewrite code.

Developer productivity and efficiency have been the key selling points for providers of AI coding tools, and research does highlight positive signs on both these fronts.

Stack Overflow’s 2025 Developer Survey found 84% of devs are currently using AI tools on a daily basis, or plan to use them. But as with other tech industry professions, the increased integration of the technology has given rise to concerns about job losses.

It’s a topic that’s hit headlines with increasing frequency over the last couple months, especially when big tech layoffs hit staff in these domains, as with the recent Microsoft cuts.

However, Skrygan tells ITPro the reality won’t quite be mass job losses and the destruction of the profession, but instead an evolution of what constitutes a ‘developer’ in this new era.

“I don’t believe in mass layoffs,” he says. “There are some layoffs, let’s be honest, some companies are laying off people – but they were laying off even before the AI revolution.”

“I look at this also in the prism of historical examples,” he adds. “So I think after the first, second, third, and fourth industrial revolution, the amount of professionals from the previous arc in absolute numbers did not decrease.”

“It stayed on the same level. But on top of that, we gained a whole new lot of new professions.”

Skrygan admits that the growth of engineering jobs “may not be as high as previously”, but this also won’t result in a drought with regard to talent. Instead, new roles aimed specifically at monitoring the use of AI and its role in the broader development lifecycle will emerge.

“Maybe another sort of profession where people are assessing what AI agents have done,” he says. “It's not about prototyping, not about algorithms, it's about high level assessment of the whole program.”

The JetBrains chief executive isn’t alone in this optimistic outlook, either. In a recent blog post, GitHub CEO Thomas Dohmke suggested the evolution of developers will see them move from “writing code to architecting and verifying the implementation” of work conducted by AI agents.

His comments come hot on the heels of a podcast appearance in June where he also dispelled the myth of mass layoffs, arguing that the technology will be a net positive for the industry.

Skrygan agrees with this argument, adding that AI also has the potential to “democratize software development” for those perhaps without the technical expertise.

“This is about a new level of creators. Maybe some of them will be just creators and will use no-code,” he says.

Upskilling and evolution

Regardless of what shape the developer or engineer role will take in years to come, one thing is certain: practitioners across the industry will have to adapt to rapid changes, necessitating upskilling on both an enterprise and individual level.

Research from Gartner in late 2024 highlighted the need for rapid upskilling to compensate for changes brought about by AI, especially autonomous agentic AI, which is expected to result in a new wave of “AI-native software engineering”.

Speaking at the time, Philip Walsh, senior principal analyst at the consultancy, said AI will demand a “new breed of software professional, the AI engineer”.

Meeting demand for this will require an increase in the number of developers skilled in the technology, according to the consultancy.

“I think what matters here is the level of the knowledge of these AI tools,” Skrygan says. “It is also tricky. It's not just about one single magical button generating code. It’s about how you prompt agents, which agent are you using? How do you verify the liability of the generated code?”

Developing skills in these areas will be a key differentiator for developers and engineers moving forward, Skrygan adds.

Cutting through the AI hype

One area Skrygan is keen to highlight is the hype surrounding generative AI and, more recently, agentic AI. It’s not a silver bullet for developers, or professionals in any line of work.

“After talking a lot with CTOs and CIOs, real technical leaders from different organizations, I feel kind of a sentiment that they have started to be a little disappointed with AI being overhyped,” he says.

“Don't get me wrong, and don't get them wrong. AI is a fabulous tool to start something new, prototype fast,” Skrygan adds. “You don't need five engineers to build an MVP. All of this is 100% legit.”

In one case, he notes, a CTO at a large enterprise mentioned they are seeing a marked improvement in code completion using AI tools, which is reducing manual toil for developers.

But the flip side of these so-called productivity and efficiency gains is that they’re forced to go back and manually review and remediate glaring issues.

“What I see on the other hand is that we spend ten-times more [time] reviewing pull requests, and then the customer satisfaction of features went down,” he says.

“Another person said that [with] so much code being generated by AI, the quality of this code is much, much lower,” Skrygan adds. “So what we get is just another pile of technological debt, and we need to handle this.”

These examples are just a glimpse into the delicate balancing act development teams now face when using AI. It’s delivering improvements, but these are being offset by having to solve new problems created by the technology itself.

A study from Harness in late 2024 found that while 92% of respondents used AI tools to increase the volume of code shipped into production, 67% end up spending more time debugging code than previously.

Over two-thirds (68%) also reported spending more time resolving security vulnerabilities since adopting AI tools than prior to their adoption.

This was another key issue highlighted by Stack Overflow’s recent Developer Survey. Devs are using these tools, but nearly half (46%) said they “don’t trust the accuracy” of the output from AI.

Ultimately, this comes down to how the tools have been marketed by providers. It’s within their interest to tout the benefits, but somewhere along the way the messaging has framed AI tools as a one-size-fits-all wonder tool for the enterprise.

“I think the industry is a little bit kind of fed up with phrases that will be like ‘by the end of this year, there will be no code generated by people’ or something like that,” he says.

“This is not the case. A year ago they were saying only about AGI, AGI, AGI’, and now we see that this is also not the case anymore,” Skrygan adds. “There are lots of details. It's not as easy as we expected this to be in 2023 or 2024, so we need to dig deeper.”

Ross Kelly
News and Analysis Editor

Ross Kelly is ITPro's News & Analysis Editor, responsible for leading the brand's news output and in-depth reporting on the latest stories from across the business technology landscape. Ross was previously a Staff Writer, during which time he developed a keen interest in cyber security, business leadership, and emerging technologies.

He graduated from Edinburgh Napier University in 2016 with a BA (Hons) in Journalism, and joined ITPro in 2022 after four years working in technology conference research.

For news pitches, you can contact Ross at ross.kelly@futurenet.com, or on Twitter and LinkedIn.