Engineers must relearn daily workflows as AI reshapes their roles

KUALA LUMPUR, 25 AUGUST 2026 —

The artificial-intelligence boom is rewriting engineering careers, forcing even seasoned professionals to relearn daily workflows and making entry-level roles harder to secure.

“Employers tell engineers that they must be adaptable when they graduate, but we don’t explain what that looks like, demonstrate it, or help them develop it,” Brunhaver told IEEE Spectrum. Her ongoing interviews with engineering managers, early-career engineers and undergraduates show that definitions of adaptability vary widely by employer and industry.

In software engineering, the daily toolkit can change overnight as libraries and frameworks evolve. Aerospace or biomedical engineers meanwhile face shifting regulations and certification requirements. Brunhaver defines adaptability simply as “the ability to recognise that change or uncertainty is occurring and then respond effectively.”

The pressure is intensifying. Technology, media and telecom sectors are experiencing the fastest pace of skill turnover, according to a June 2026 report by PwC. The World Economic Forum’s 2025 Future of Jobs Report projects that employers across all sectors expect the core skills of 39% of workers to change by 2030.

For engineers already in the workforce, the transition can feel chaotic. Jenna Butler, a Microsoft research scientist studying developer well-being, likens the current phase to the “uncomfortable middle” of a shift that has happened before but is moving faster. Software teams are now spending far more time reviewing AI-generated code, she says, leading to conflicting workflows.

“If you ask 20 developers, you get 23 different ways of working with it,” Butler told IEEE Spectrum. “Everyone is trying to sort it out.” Yet she argues that the underlying problem-solving and communication tasks of engineering remain stable even as tools change.

Andy Hunt, co-author of the 1999 software classic The Pragmatic Programmer and author of its 2019 and 2026 editions, agrees. He notes that the book’s advice about systems thinking and steady problem-solving still holds today. Hunt cautions engineers against defining themselves by a single language or tool.

“Identifying as a Java programmer is like a carpenter saying ‘I’m a hammer user’,” Hunt said. Yet that narrow focus is often reinforced by hiring practices that filter résumés for specific languages or years of experience.

This mismatch complicates adaptability. Butler advocates protected learning time—even one hour per week—where engineers can explore new tools without pressure to produce immediate code. “Organisations won’t get sudden change in their people if they don’t give them time and space to learn how to work differently,” she said.

Brunhaver urges educators to embed real-world variety into curricula: internships, team projects, community service and leadership roles. Reflection and metacognition also matter, she says. “To adapt, you need to believe you have agency and the ability to get through a situation.”

The stakes are high. Engineers who resist retraining risk falling behind, while those who embrace change shape the future of their fields. Butler stresses that professionals have a choice in how they use AI and which models they adopt. “We must be adaptable, but we also direct that flow,” she said. “The future with AI is not predetermined.”

Related: Petronas