Two Labor Markets in One Industry

The most disorienting feature of the 2026 tech labor market is that two completely contradictory things are true at the same time. Through early 2026, over 51,000 technology workers were laid off across more than 130 companies, according to layoff trackers. Amazon alone accounted for an estimated 16,000 of those positions, representing over half of all announced tech layoffs in January. Meta cut approximately 1,500 jobs from its Reality Labs division. Autodesk and Salesforce each trimmed roughly 1,000 positions. If current trends continue through the year, RationalFX projects the industry could see over 270,000 total tech job losses in 2026, far outpacing the approximately 124,000 cuts recorded across 269 companies in 2025.

And yet, in those same weeks, companies posted tens of thousands of new positions they could not fill. AI engineers, machine learning researchers, cloud architects, cybersecurity specialists, and data platform engineers remained desperately sought-after. AI engineer salaries jumped to an average of $206,000 in 2025, a $50,000 increase from the prior year. At top firms like NVIDIA, Google, Apple, and Meta, machine learning engineers command between $230,000 and $362,000. Specialists in generative AI and LLM fine-tuning earn premiums of 40 to 60 percent above baseline ML salaries. And still, companies cannot fill the roles. The global demand-to-supply ratio for AI talent sits at roughly 3.2 to 1: approximately 1.6 million open positions against only 518,000 qualified candidates available.

This is not a temporary anomaly. It is the defining structural feature of the technology labor market in 2026: simultaneous mass layoffs and mass talent shortages, existing in the same industry, sometimes in the same company, occasionally in the same week.

Survey data from late 2025 crystallized the paradox. Roughly 55 percent of hiring managers at technology companies reported expecting layoffs at their organization in the near future. At the same time, 44 percent said that AI-related capabilities were the primary driver of both the cuts and the new hiring. Companies are not simply shrinking. They are reshaping their workforces at speed, eliminating roles that do not align with their AI-driven strategies while frantically trying to hire for roles that do.

Where the Cuts Are Falling

Understanding the paradox requires looking at where, specifically, the layoffs are concentrated. The cuts are not random. They follow a pattern that reveals the industry’s strategic logic.

The heaviest cuts have fallen on roles related to traditional software support and maintenance. Quality assurance engineers, manual testers, and technical writers have been among the most affected. Companies are betting, sometimes correctly and sometimes prematurely, that AI tools can handle much of the work these roles traditionally performed. Automated testing, AI-generated documentation, and code review bots have matured enough that some companies feel confident reducing their human investment in these areas.

Corporate function roles have been hit hard. HR generalists, internal communications specialists, financial analysts, and operations coordinators within tech companies have faced significant cuts. Customer service, content moderation, and data entry roles face particular pressure. The logic is consistent: AI tools for data analysis, employee self-service, and operational coordination are replacing the manual work that these roles centered on.

Program management and project management have seen notable reductions, particularly at companies that have flattened their management hierarchies. Amazon specifically targeted middle management positions in its January restructuring, consolidating teams to reduce organizational complexity. When AI-powered project tracking tools can generate status reports, flag risks, and coordinate timelines, the project manager’s information-aggregation function becomes redundant. Strategic program management remains valuable, but the number of people required has decreased.

Sales development representatives (SDRs) and early-stage sales roles have been cut at companies that have deployed AI-powered outbound tools. Companies such as Block, Autodesk, Ocado, and Pinterest have trimmed go-to-market teams and overlapping product functions. Salesforce is reducing headcount across a mix of sales, marketing, and support roles as part of an organizational simplification effort. The high-volume, low-personalization outreach that traditionally occupied SDR teams is being automated, with human salespeople reserved for higher-value, relationship-based selling later in the funnel.

Notably, the cuts are not limited to junior or entry-level positions. Mid-career professionals with 8 to 15 years of experience in non-AI domains are finding themselves particularly vulnerable. They are too senior (and expensive) to retrain easily, but their existing skills do not map to the roles companies are hiring for. AI was the cause of nearly 55,000 layoffs in the US in 2025 alone, and organizations are increasingly choosing not to replace vacated roles, instead exploring how AI can fulfill those functions.

Where the Hiring Is Desperate

On the other side of the paradox, companies are spending aggressively and still failing to fill positions in several key areas.

AI and machine learning engineering remains the most constrained talent market. The demand for engineers who can build, fine-tune, and deploy production AI systems continues to outstrip supply by a wide margin. This includes not just model-building roles but the entire ecosystem: MLOps engineers who manage model deployment (commanding 20 to 35 percent salary premiums), AI infrastructure engineers who build the platforms that support model training, and applied AI engineers who integrate models into products. Large Language Model specialists earn 25 to 40 percent more than general ML engineers. The job outlook projects 40 percent growth in AI specialist roles through 2030, but the talent pipeline forecasts only 2.1 million qualified professionals against 4.2 million roles needed.

Cloud and infrastructure engineering is another area of acute shortage, particularly for engineers with experience in multi-cloud architectures and Kubernetes-based platforms. The migration to cloud continues to accelerate, and the complexity of managing cloud environments at scale requires specialized expertise that takes years to develop.

Cybersecurity is experiencing what may be the most severe talent shortage in any technology discipline. The ISC2 2025 Cybersecurity Workforce Study, based on over 16,000 participants worldwide, found that global demand for cybersecurity professionals reached 10.2 million, with a current workforce of only 5.5 million, leaving a gap of 4.76 million unfilled roles. For the first time, ISC2 declined to publish its traditional workforce gap estimate, noting that respondents now view skills shortages as eclipsing headcount shortages alone. Nearly 59 percent of cybersecurity teams reported critical or significant skills gaps, up from 44 percent the prior year. AI security and cloud security top the most-needed skills list, with 41 percent and 36 percent of respondents citing them as vital, respectively. The consequences are real: 88 percent of organizations reported experiencing at least one significant cybersecurity incident in the past year because of skills shortages. Security engineers with expertise in these areas command compensation premiums of 20 to 40 percent over comparable non-security roles.

Data engineering and analytics roles remain strong, particularly for professionals who can work with modern data stack technologies and support AI-driven analytics. The data infrastructure that underpins AI systems requires specialized engineering, and companies are investing heavily.

The common thread across all these growth areas is that they require deep, specialized expertise that cannot be acquired quickly. More than 90 percent of organizations say IT skills shortages will affect them by 2026, with an estimated $5.5 trillion in lost global productivity tied to the gap.

Advertisement

The Structural Mismatch Problem

The tech layoff paradox is fundamentally a skills mismatch problem, but it is a mismatch of a specific and particularly intractable kind.

In a normal labor market transition, workers displaced from declining sectors can retrain for growing sectors over a period of months to a few years. The auto worker learns to install solar panels. The coal miner retrains as a wind turbine technician. The transition is difficult but feasible because the skill distance is manageable.

The current tech labor market transition is different because the skill distances are large and the timelines are compressed. An experienced QA engineer or project manager cannot retrain into a machine learning engineer or cybersecurity architect in less than two to three years of intensive study and practice. These are not adjacent skills. They are different disciplines with different foundations, different ways of thinking, and different experience requirements.

Moreover, the demand side is moving faster than the supply side can respond. By the time a displaced tech worker completes a retraining program in AI or cybersecurity, the specific skills in demand may have shifted again. The technology landscape is evolving at a pace that makes traditional retraining models, which assume relatively stable target skill sets, increasingly inadequate.

This creates a painful dynamic. Companies cannot find the specialized talent they need. Workers with experience in declining areas cannot bridge the gap to growing areas quickly enough. And the usual market mechanisms (rising wages in scarce fields attracting workers from surplus fields) work too slowly to resolve the imbalance. Despite AI engineers earning $206,000 on average and top-tier specialists pulling $362,000, 35 percent of companies still cite high salary expectations as their top recruitment challenge, suggesting even these premiums are not producing enough qualified candidates.

The geographic dimension compounds the problem. Many of the workers being laid off are concentrated in specific metro areas (Seattle, the Bay Area, Austin) and have housing costs and family situations that make relocation difficult. Meanwhile, the demand for specialized talent is often in the same areas, but the demand is for different people with different skills, not for the people who were just let go.

What Workers Should Do: A Strategic Response

For technology workers navigating the paradox, whether they have been laid off or are watching anxiously from employed positions, the situation demands strategic thinking rather than reactive panic. Several principles should guide career decisions.

The first principle is to invest in AI adjacency even if you are not an AI specialist. You do not need to become a machine learning researcher to be valuable in an AI-driven organization. What you need is to understand AI well enough to apply it within your existing domain. A product manager who understands how to define and scope AI features is dramatically more valuable than one who does not. A QA engineer who can design testing frameworks for AI systems is in demand. A technical writer who can document AI APIs and create user guides for AI-powered products fills a growing niche. The key is not to abandon your existing expertise but to augment it with AI literacy.

The second principle is to develop T-shaped expertise. The most resilient tech professionals have deep expertise in one area (the vertical bar of the T) combined with working knowledge across several adjacent areas (the horizontal bar). In the current market, the ideal vertical expertise is in one of the growth areas (AI, cloud, cybersecurity, data engineering). But even without deep specialization in these areas, a broad horizontal bar that includes AI fundamentals, cloud concepts, and security awareness makes any tech professional significantly more employable.

The third principle is to build proof of capability rather than collecting credentials. The market increasingly values demonstrated ability over certifications and degrees. Contributing to open-source AI projects, building and deploying personal AI projects, writing technical content that demonstrates understanding, and participating in community discussions about AI applications, these tangible artifacts are more persuasive to hiring managers than a line on a resume listing a completed online course.

The fourth principle is to consider adjacent industries. The demand for technology talent extends far beyond the technology industry itself. Financial services, healthcare, manufacturing, energy, and government are all hiring aggressively for technology roles, and they often face less competition from other employers. The global net employment outlook for early 2026 stands at +24 percent, with 40 percent of employers planning to increase headcount, according to ManpowerGroup data. A cybersecurity professional who works in healthcare may command slightly lower compensation than one at a Big Tech company but faces dramatically better job security and less competitive hiring.

The Path Forward: What Needs to Change

The tech layoff paradox exposes a fundamental inadequacy in how the technology industry manages its workforce transitions. Several changes could mitigate the worst effects.

Companies that lay off workers should invest in meaningful transition support, not just severance checks and outplacement services. The most responsible companies are beginning to offer structured retraining programs that help displaced workers bridge into high-demand areas. This is not pure altruism; it builds goodwill, protects employer brand, and contributes to a healthier talent ecosystem that the company will draw from in the future.

The education system needs to respond faster. Traditional university programs, with their multi-year curriculum development cycles, cannot keep pace with the speed of change in technology. Shorter-form, intensive programs, partnerships between employers and educational institutions, and stackable credential systems that allow workers to build expertise incrementally are all part of the solution. The skills-based hiring movement, which is gaining traction in parallel, creates a natural ecosystem for these alternative credentials.

Government workforce development programs need to update their assumptions. Most existing programs are designed for manufacturing-to-services transitions, not for within-industry skill shifts. Programs that specifically address the challenge of helping experienced technology workers retrain for adjacent technology disciplines would address a significant gap. The fact that more than 25 US states have now dropped degree requirements for government tech jobs signals an awareness of the problem, but policy needs to go further into active retraining support.

Most fundamentally, the industry needs to stop pretending that the layoff paradox is a temporary misalignment that will self-correct. It is a structural feature of an industry undergoing rapid technological transformation. The AI talent shortage alone is projected to persist through 2030, with demand roughly double the forecasted supply. Managing it effectively requires institutional responses at the company, educational, and policy levels that match the scale and permanence of the challenge.

Advertisement

🧭 Decision Radar (Algeria Lens)

Dimension Assessment
Relevance for Algeria High — Algeria’s tech sector is growing (IT services market at $1.9 billion in 2025, 500+ digital projects planned for 2025-2026), but the country faces its own version of the mismatch: abundant graduates but acute shortages in AI, cybersecurity, and cloud skills. Youth unemployment exceeds 30% despite nearly 2 million university students.
Infrastructure Ready? Partial — Algeria’s 2030 Digital Transformation Strategy and Algerie Telecom’s 1.5 billion dinar investment in AI/cybersecurity startups signal intent, but enterprise-scale cloud infrastructure, MLOps platforms, and cybersecurity operations centers are still nascent.
Skills Available? No — Algeria has strong STEM enrollment but limited pipelines for the specific high-demand skills (ML engineering, cloud architecture, AI security) that define the global hiring side of the paradox. The 18 centers of excellence are a starting point, not yet at scale.
Action Timeline Immediate to 12 months — Algerian tech workers and students should begin AI/cybersecurity upskilling now. The global talent shortage creates remote work opportunities for Algerian professionals who can bridge the skills gap, especially given the favorable CET time zone for European clients.
Key Stakeholders Ministry of Knowledge Economy, Startups & Micro-Enterprises; Algerian Startup Fund; Algerie Telecom’s incubation programs; universities with CS/engineering departments; individual tech professionals building AI-adjacent skills.
Decision Type Strategic — Both individual career decisions and institutional workforce development policy.

Quick Take: Algeria sits in a unique position. The global layoff paradox creates an opening: while Western companies cut traditional roles and cannot fill AI/cybersecurity positions, Algerian tech professionals who invest in these high-demand skills can access remote opportunities with global employers. The $1.9 billion domestic IT market and 500+ planned digital projects also need these same skills locally. The window is open, but it requires proactive skill-building. Algeria’s graduates should treat the global skills mismatch not as a distant problem but as their most immediate career opportunity.

Sources & Further Reading