AI Is Reshaping Engineering Jobs. Here Is What Actually Keeps You Employable

Key points

• The World Economic Forum predicted in 2025 that 39 percent of workers' core skills will change by 2030 across all industries.
• Technology, media and telecoms jobs are seeing the fastest pace of skill turnover, according to a June 2026 PwC report on AI's effects.
• Arizona State University professor Samantha Brunhaver received a National Science Foundation award in 2020 to study workplace adaptability in young engineers.
• Microsoft research scientist Jenna Butler calls the current AI transition an "uncomfortable middle" period that is moving faster than historical shifts.
• Problem-solving and communication skills outlast any specific coding language or tool.

AI has not just changed what engineers build. It has changed what they are expected to know, how fast they must learn it, and whether they get hired at all. That tension connects directly to what we reported on 13 August when the US labour market shed 23,000 jobs just as AI tools were spreading across industries.

The honest answer from researchers and practitioners, reported by IEEE Spectrum AI, is that specific tools matter less than the ability to swap them out without panicking.

What does "be adaptable" actually mean?

Adaptability means spotting when something around you is changing and responding usefully rather than freezing. That sounds simple. In practice, almost nobody teaches it.

Samantha Brunhaver, an associate professor of engineering at Arizona State University, has spent years interviewing engineering managers and early-career employees about this gap. Her finding: everyone tells young engineers to be adaptable, and almost nobody shows them how.

"We tell engineers that they need to be adaptable when they graduate, but we don't actually explain what that means, demonstrate what that looks like, or help make sure that they're developing it," Brunhaver says.

Her research breaks adaptability into three concrete steps: notice that something is changing, weigh your options, then act. That cycle applies to a software engineer whose AI coding assistant changed overnight just as readily as it applies to an aerospace engineer tracking a new safety regulation.

Is this AI shift really different from past changes?

Yes and no. Jenna Butler, a research scientist at Microsoft who studies how developers work and stay healthy at work, says big technology shifts always produce a chaotic in-between period before a new normal settles. AI is one of those shifts. The difference is pace.

"I think we're still in this in-between, difficult period that we've seen before, but it is maybe moving faster than it has historically," Butler says.

For software engineers specifically, AI tools are generating large amounts of code that someone still has to check. That has dramatically increased the volume of code review, the process of reading and approving AI-generated output before it goes live, landing on developers' plates even as the tools are supposed to save them time.

What skills actually survive?

Andy Hunt co-wrote "The Pragmatic Programmer" in 1999, a book still taught in computer science courses today. Working on its 20th anniversary edition, he was struck by how little the core advice had aged.

His point: coding languages are tools, not identities. Calling yourself a Java programmer, he says, is like a carpenter claiming they specialise in cordless drills. The drill is a means, not a mastery.

"The fundamental part of the job is problem solving and communication, and that's always going to be there," Hunt says.

Butler adds a practical prediction: software engineering will shift toward guiding AI models through complex problems rather than writing every line manually. Engineers who went into the field because they enjoy solving problems will find plenty to enjoy. Those who loved coding purely as a craft may find the transition harder.

Hunt goes further, arguing that employers who filter candidates by specific language keywords are looking at the wrong thing. Understanding systems thinking and constraints, he says, is still the hard part, and always has been.

What should employers and educators do?

Brunhaver argues that universities need to offer varied real-world experiences, including internships, team projects and leadership roles, so students practise adapting before their first job depends on it. Encouraging reflection matters too: metacognition, the habit of thinking about how you think, helps people deploy adaptability more deliberately.

For managers, Butler's advice is specific: carve out one hour a week for engineers to learn new tools, with no deliverable attached. Not a training module tied to a deadline. Just time to explore.

"You're not going to get this sudden change in your people if they don't have time and space to learn how to work differently," Butler says.

Without that space, engineers under pressure to be more productive tend to stick with what they already know. Burnout follows. That pattern is worth watching: the tools are accelerating faster than most organisations are building the slack to absorb them.

Common questions

Do I need to learn a new coding language to stay relevant?

Not urgently. The ability to learn new tools quickly matters more than which tool you currently know. Problem-solving and communication skills transfer across languages and platforms.

Who is responsible for helping engineers adapt, workers or their employers?

Both. Individuals should take initiative, but employers who do not create time and support for learning are likely to see their teams burn out or fall behind rather than successfully adapt.