Main image of article Demand for AI Expertise Grows as IT Job Postings Fall

Demand for AI skills is reshaping the tech hiring market, even as overall recruiting remains muted.

New labor data shows employers are pulling back on broad tech hiring while concentrating heavily on candidates with AI expertise—a shift that’s redefining which roles get approved and which stay frozen.

Tech postings remain down compared to last year, but listings requiring AI capabilities have climbed sharply and now appear in more than half of all U.S. tech openings, according to Dice’s November Jobs Report.

CompTIA’s September Tech Jobs Report, based on information from the U.S. Bureau of Labor Statistics, also found a sharp upward trend in its AI Hiring Intent Index, with hiring for dedicated AI job titles on the rise in major IT job markets including San Jose, New York and San Francisco.

Marcus Fontoura, CTO for Azure Core, Microsoft, explains at the top end, companies are hiring aggressively for talent that can move the AI needle.

“Everywhere else, they’re cautious,” he says. “It’s still too early to understand the impact of AI in many sectors. AI will primarily make humans more productive. We are not thinking that AI will replace humans.”

His view in tech is that AI is helping us increase the quality of what must be done, for example junior developers tackling more complex tasks.

“Think about the way word processors and spell correctors made journalists and writers more productive,” he says. “We are continuing pushing the productivity frontier with technology, and AI will help us do that.”

Ali Gohar, chief human resources officer at Software Finder, says what’s happening here isn’t a contradiction; it’s a restructuring.

“Many businesses are starting to realize that they need fewer people in general information technology roles and are focusing more on the support side of things,” Gohar explains.

The additional personnel are being applied to the more specialized work involving AI and automation.

He says companies must begin to optimize the conventional staff they employ and heavily invest in AI personnel to be competitive in the new technologies being developed.

“Tech job shortages are just evolving,” he adds.

Michael Morris, head of Torc by Randstad Digital, explains the stagnancy in headcount we’re seeing is influenced by macroeconomic uncertainty but also likely reflects a strategic recalibration that is driven by the widespread adoption of AI and investment into AI capabilities.

“We are seeing in some of the restructuring announcements an explanation that savings realized by AI investments that have been made or planned,” he says.

Many companies are no longer hiring in bulk but are instead focusing on high-leverage talent segments, proactively building resilient talent pipelines through smart hiring tools, AI-powered platforms, and structured training pathways.

At the same time, talent itself is looking for career acceleration through meaningful learning journeys, and companies that align with this aspiration are seeing higher productivity and lower attrition.

“We feel companies aggressively looking to train and mentor their talent on new skills will have a strong long-term benefit when it comes to attracting and retaining top talent across their organizations,” Morris says.

Tim Burke, founder, president and CEO of Quest Technology Management, says entry-level technologists often lack practical experience in data engineering, model operations, security-aware development, and the ability to tie AI work to real business outcomes.

“Organizations don’t need multi-year roadmaps to close these gaps,” he says. “They need focused, hands-on experience.”

He explains short, project-based learning cycles structured his ability to tie AI work to real business outcomes.

Mentorships and pairing junior staff with senior engineers on real deployments can accelerate readiness far faster than traditional training programs.

“The key is exposing them to production-grade systems early, so they learn the realities of scale, governance, and reliability,” he says.

Fontoura says today’s engineering talent needs stronger systems thinking, adaptability, and a deeper understanding of data.

He points to the growing complexity of modern technology environments and stresses the importance of modeling that complexity effectively.

“We live in a complex world, and we need to model a complex world using technology. Understanding feedback loops, externalities, incentives, and failure modes matters.”

A growth mindset is equally important for long-term performance and resilience on engineering teams. As Fontoura puts it, “we need to hire people that can learn from their mistakes, do well with feedback, and are interested in growing.”

He highlights a persistent gap among early-career technologists around data literacy and practical judgment.

“Many juniors can code, but don’t reason about data quality, drift, bias, or observability,” he says.

Fontoura says in the long run, with automation of procedural and mechanical tasks, we’ll be able to have flatter organizations.

“This means that we’ll need less middle management,” he says.

To handle an organization of 100 people, there will be a need for just a few managers as most management tasks will be automated.

“We can therefore have more individual contributors directly writing code, which is great,” Fontoura says. “Less management overhead and more value produced.”

Morris says early career professionals should focus on building AI literacy, pursuing targeted upskilling or reskilling opportunities, and focus on building human-centric skills like critical thinking and communication to stay competitive as traditional career ladders shift.

He adds individuals can influence their environment by fostering creativity and curiosity, understanding their own emotional intelligence and thinking using ethical judgement in a world where AI agents are given more leeway to execute complex business processes.

“At the end of the day, adaptability will be the underlying skill professionals can build their career path upon,” he says.