Estimated reading time: 5 minutes
Why are tech companies struggling to fill positions, despite a well of talent available in the industry?
When the 24-year-old AI researcher Matt Deitke accepted a $250 million job offer from Meta earlier this summer, the deal made international headlines. It also heralded a new reality: after years of nonstop tech layoffs, technologists are suddenly pocketing paychecks on par with NBA stars.
The other side of that coin, however, is that employers are finding there simply aren’t enough qualified workers to go around – all thanks to the AI boom. This spring, McKinsey reported that a fast-widening tech talent gap is poised to balloon from 1.4 million to 3.9 million workers by 2027 within the European Union alone.
How did tech go from being an industry that couldn’t shed its workers fast enough to one that’s desperate to hire? Experts tell LeadDev that the problem largely boils down to unrealistic hiring expectations and a misalignment between job specs and available talent, not an actual lack of skilled people.
Your inbox, upgraded.
Receive weekly engineering insights to level up your leadership approach.
What’s really happening with the tech talent pool?
There’s no question that the demand for AI job skills is rising, and it’s happening well beyond tech.
“Our clients in healthcare, finance, manufacturing, logistics, and even defense are rapidly hiring for machine-learning engineers, prompt engineers, AI ops specialists, and data scientists,” says Bill Peppler, the Orlando, Florida-based COO for the professional-staffing agency Kavaliro. The AI job boom has also increased the need for workers with AI-adjacent skillsets, such as cloud engineers and data platform engineers to support AI pipelines, as well as cybersecurity roles.
Often, companies struggle to find viable candidates due to the newness of the technology in question. “The job descriptions have evolved faster than the skills of the available workforce. Someone who worked on Natural Language Processing models two years ago might not have hands-on experience with LLMs, Reinforcement Learning from Human Feedback, or prompt chaining,” Peppler says. “Also, talent acquisition departments are not always up to speed on these newer technologies, so the job descriptions can be difficult to get right and hurt a company’s ability to find talent.”
Lei Gao, the Hong Kong-based CTO for the AI social commerce platform SleekFlow, contends that the challenge many companies currently face is less of a talent shortage than a skill gap.
“Everyone wants to roll out LLMs or create AI copilots, but there’s a fairly limited talent pool of engineers who know how to take those models out of the research lab and turn them into a production-quality system,” Gao points out. Even fewer can operate at scale while accounting for reliability, latency, and data privacy. The reason why has nothing to do with a lack of talent, Gao says, but because turning AI into production-ready systems is challenging. It’s also still fairly uncharted territory.
In short, transformative new tech systems come with a learning curve for the workforce.
Though the newness of AI-related skillsets has left little time for the workforce to build capacity, many organizations are holding out for ideal candidates. “It really is often a case of companies chasing unicorns and then blaming the market when they can’t find one,” says Nick Derham, Director at Adria Solutions, an IT and digital recruitment agency in the UK.
Much of this behavior comes down to fear: given the tech-market chaos of the last few years, businesses want to “play it safe” by hiring people who’ve already done more or less the exact same role, Derham says. “The problem is that, with fast-moving areas like AI, that kind of experience is rare. And by the time someone has it, they’re already being snapped up by the biggest players.”
At the same time, Derham says his firm is constantly speaking with highly-skilled developers, engineers, and data professionals who are having trouble landing roles – and not just in AI. “Areas like DevOps, data engineering, and software roles now come with much longer lists of ‘must-haves’ than they did even a year ago.” Continued downsizing across the industry has left companies unwilling to take any chances.
Gao agrees that the standards for these kinds of roles have increased, especially in startup environments where one hire can have a tremendous impact on the eventual success of the company. However, unlike tech giants such as Meta, smaller companies and startups aren’t typically able to drop hundreds of millions of dollars to poach top AI talent. They need to find their own ways of standing out.
More like this
How companies – and candidates – can reinvent the hiring wheel
In the high-stakes matchmaking game between tech-industry job seekers and the companies looking to hire, both parties need to reconsider how they position themselves to attract the other’s attention. For startups and smaller organizations, that means focusing on providing efficient interviewing and onboarding processes, employee flexibility, and clear opportunities for growth. In other words, emphasize the things money can’t buy.
“At SleekFlow, we can’t always keep up with Big Tech on compensation, but we can provide quicker decision-making, actual ownership, and significant problems to solve,” Gao says. “Hiring today is about finding individuals who can excel in ambiguity, work across functions, and grow with the organization. We’ve learned that emphasizing mission alignment, internal mobility in the company, and global hiring to get the right people on board works. It requires more effort, but it works.”
November 3 & 4, 2025
LeadDev Berlin agenda is live! 🙌
Peppler echoes Gao’s perspective that smaller organizations can get ahead by offering intangibles that may be more difficult to obtain at a major company. “The startups we work with have benefited from attracting candidates who care more about innovation and impact than comp alone,” he says.
Organizations could also stand to rethink their hiring processes, as well as how they evaluate talent.
Peppler advises engineering leaders to make sure they are involved in creating job descriptions and are communicative with recruiters about which skills are must-haves versus good-to-haves for the position. He also recommends that hiring managers prioritize candidates who show adaptability and learning velocity, and don’t simply evaluate potential hires on the basis of their past roles.
Derham offers similar wisdom. “Be clear on what you need, but be flexible on how it shows up,” he says. “Very few candidates will tick every box, but many can grow into the role quickly if given the chance.”
As for how recently laid-off engineers should approach this strange new landscape, the advice is simple: keep learning as broadly as possible.
“Don’t pursue just the most trendy buzzwords,” says Gao. “Develop depth in systems thinking, AI fundamentals, and production-level skills. Companies require builders to scale responsibly, not prototype hastily. Opportunity lies ahead, but fitting into the correct org counts.”
