Three announcements in one week tell you where the tech job market is heading. Meta is cutting 8,000 workers to fund a $135 billion AI buildout. Microsoft is offering voluntary buyouts to up to 7% of its U.S. workforce — a first in the company’s 51-year history. And Snap fired 1,000 people after its CEO noted that AI now generates more than 65% of the company’s new code. The common move: spend less on people, spend more on compute.
Through late April, more than 100,000 tech workers have lost their jobs in 2026. Nearly half of those cuts cite AI as a direct factor. The people building AI are doing fine. Everyone else is watching the floor shift.
This Week’s Layoffs
Meta: 8,000 Gone, $135 Billion Redirected
Meta told employees on April 23 that it will eliminate 8,000 positions — 10% of its 78,865-person workforce — effective May 20, with another 6,000 open roles scrapped. The company’s 2026 capital expenditure guidance sits between $115 billion and $135 billion, nearly double what it spent in 2025.
Chief people officer Janelle Gale framed the cuts as necessary “to offset the other investments we’re making.” Wedbush analyst Dan Ives was more direct: Meta is automating tasks that once required large teams.
U.S. employees get 16 weeks of base pay plus two weeks per year of employment. European employees get comparable packages. The layoffs start May 20.
Microsoft: Buyouts for the First Time Ever
On the same day Meta announced its cuts, Microsoft offered voluntary buyouts to eligible U.S. employees — the first such program in the company’s history. About 7% of the U.S. workforce qualifies: employees at senior director level and below whose combined age and years of service total 70 or more.
That potentially affects up to 8,750 people. Details go out May 7. Microsoft is expected to spend $145 billion in capital expenditure this fiscal year, most of it on AI data centers. The company chose buyouts over layoffs — but the direction is the same.
Snap: AI Writes the Code, Humans Get the Layoff Notice
Snap cut 1,000 employees on April 15 — 16% of its workforce. CEO Evan Spiegel told staff that AI now generates more than 65% of Snap’s new code and described the restructuring as moving to “a new way of working” with smaller, AI-augmented teams.
The restructuring will save Snap over $500 million annually by the second half of 2026. Its stock jumped 8% on the announcement.
UKG: 950 Cuts, AI as the Stated Reason
HR software company UKG laid off 950 employees on April 15, citing “rapidly evolving market shifts — including changes in technology driven by AI.” This was UKG’s second major cut in 18 months, following 2,200 layoffs in late 2024.
The Numbers So Far
The tech layoff trackers are converging on a grim picture for 2026:
- 100,000+ tech workers laid off through late April, across 155+ layoff events
- ~48% of cuts explicitly cite AI as a factor — either automating roles or redirecting budget toward AI engineering
- $650 billion in combined capital expenditure planned by Amazon, Google, Meta, and Microsoft for 2026
- At the current pace, analysts project the full-year total could approach 265,000 job losses globally
The largest single event was Oracle’s 30,000-person cut. Meta’s 8,000, Microsoft’s potential 8,750, and Snap’s 1,000 are the most recent major additions. Tom’s Hardware reports that Q1 alone saw nearly 80,000 tech workers lose their jobs.
Who’s Getting Hired
The flip side of the layoff numbers: AI-specific roles are booming.
Mid-level machine learning engineers now earn between $149,000 and $192,000 nationally, with salaries up 9% year-over-year — one of the largest jumps in tech. Senior ML engineers command $160,000 to $226,000 in base pay, with total compensation regularly clearing $300,000 at Big Tech companies. In San Francisco, mid-career AI engineers pull $187,000 to $220,000 before equity.
The most in-demand roles: AI/ML engineers, MLOps engineers, forward-deployed engineers, AI governance and ethics specialists, and data annotators. Demand for AI governance skills is up 150%. AI ethics expertise is up 125%.
LinkedIn data shows 1.3 million new AI-related jobs created globally in the past two years. The number of workers in roles requiring AI fluency has grown sevenfold since 2023 — from roughly 1 million to 7 million.
But those numbers require context. The jobs being created require different skills than the jobs being eliminated. A laid-off marketing coordinator doesn’t become an ML engineer overnight. The labor market is bifurcating, not rebalancing.
The Junior Developer Crisis
The most consequential shift may be happening at the entry level. Entry-level tech job postings have dropped roughly 67% between 2023 and 2025. Employment for software developers aged 22-25 has declined nearly 20% from its late-2022 peak. Tech-specific internship postings are down 30%.
The reason is straightforward: AI coding tools now handle the tasks that junior developers were hired to do. Boilerplate code, test generation, documentation, CRUD scaffolding — the work that companies used to give new hires as a way to learn the codebase is increasingly done by Copilot, Cursor, and Claude Code.
There’s a deceptive counter-signal in the data. Job postings labeled “entry-level software engineer” grew 47% between late 2023 and late 2024. But actual hiring into those roles dropped 73% in the same window. Companies are posting the roles. They’re not filling them.
IEEE Spectrum notes that the skills employers expect from entry-level candidates have shifted dramatically. “Entry level” now means proficiency with AI-assisted development workflows, not basic programming knowledge. The bar has moved, and the pipeline that used to produce senior engineers — give juniors simple tasks, let them grow — is breaking down.
Some companies are still hiring juniors. Enterprise software vendors, financial institutions, healthcare platforms, and infrastructure companies recognize that senior engineers have to come from somewhere. But the consumer tech sector, where many developers started their careers, has largely stopped.
What This Means
The tech industry is running a real-time experiment: what happens when you simultaneously cut the people doing the work, raise the bar for new hires, and bet everything on AI tools replacing both.
The optimistic read is that AI creates more value than it destroys, the new jobs pay better than the old ones, and displaced workers retrain into higher-value roles. The LinkedIn and World Economic Forum data about 1.3 million new AI jobs supports this to a point.
The pessimistic read is that the transition is faster than people can adapt, the new jobs require fundamentally different skills, and the collapse of junior hiring creates a long-term talent pipeline problem that no one is solving. Over 90% of enterprises are projected to face critical skills shortages this year. When everyone needs senior AI engineers and nobody is training new ones, the math stops working.
The data doesn’t resolve this cleanly in either direction. What it does show is that the pace of change is accelerating. Three years ago, no Fortune 500 CEO was publicly crediting AI for making layoffs possible. This week, three of them did it in the same seven-day span.
What You Can Do
If you’re currently employed in tech:
- Learn AI-assisted development workflows now. Not someday — now. The skills employers demand are changing 66% faster in AI-exposed roles than in other positions.
- Focus on what AI can’t do yet: system design, stakeholder communication, ambiguity resolution, debugging complex production issues. These are the skills that command the premium.
- If you’re a junior developer, target companies that explicitly invest in training pipelines. Enterprise, finance, and healthcare are more likely to hire and develop you than consumer tech.
If you’ve been laid off:
- AI governance and ethics roles are growing fast and don’t always require deep ML expertise. Policy background, compliance experience, and critical thinking count.
- MLOps and deployment roles sit between traditional ops and ML engineering. If you have infrastructure experience, the bridge is shorter than you think.
- The severance packages are real: Meta offers 16 weeks plus two per year of service, Microsoft is offering buyout terms, Snap provides four months. Use the runway to retool, not just to job-search.