TL;DR

Since March 2026, we’ve published nine deep-form articles tracking the collision between AI adoption and tech employment. This guide connects them into a single picture: who’s getting laid off and why, what the salary data says across the US and EU, how AI tools are changing the work itself, and where the junior developer pipeline is headed. Each section links to the full analysis.

The 2026 Tech Job Market in One Chart

The numbers below are drawn from the individual articles linked throughout this guide. They span different time windows and sources (layoffs.fyi, BLS, Levels.fyi, BCG), so treat them as a mosaic of the market, not a single unified dataset.

180K+
Tech jobs cut in H1 2026
$725B
Big Tech AI capex committed
67%
Fewer junior dev openings

The Layoff Wave: Who, How Many, and the AI Connection

Two of the biggest stories in 2026 tech have been the scale of layoffs and the frequency with which companies cite AI as the reason.

In Q1 alone, 80,000 jobs were cut across tech, with roughly half of those companies explicitly blaming AI restructuring. Oracle, Amazon, and a dozen mid-size firms led the numbers. The pattern held a specific shape: companies didn’t say “we’re replacing humans with AI.” They said “we’re reallocating headcount toward AI infrastructure.” The jobs disappeared from one line on the org chart and reappeared on another, but the net number went down.

The geographic distribution is worth noting. US-headquartered companies accounted for the majority of announced cuts, but European subsidiaries felt them disproportionately. EMEA offices often lost a higher percentage of headcount than US headquarters, partly because US labor law makes it harder to do fast reductions and partly because European offices were seen as further from the AI-investment centers. For developers in the EU (and Cyprus specifically), this created an odd situation: the layoff announcements came from San Francisco, but the pink slips arrived in Dublin, Berlin, and Limassol.

Meta followed in May with 8,000 layoffs starting May 20, the company’s largest single cut since 2023. The framing was identical: “AI restructuring.” The affected roles skewed toward middle management and ops, not engineering, which tracks with the broader pattern across the industry. Engineering teams mostly survived intact, but their support structures shrank around them.

We pulled these threads together into a capex-focused analysis: Big Tech is trading 100K jobs for $725B in AI infrastructure. The argument is that the layoff wave isn’t reactive cost-cutting. It’s a deliberate capital reallocation. Companies are spending less on people and more on GPUs, and they’re being explicit about it in earnings calls. Microsoft, Google, Meta, and Amazon have collectively committed $725B in AI infrastructure spending through 2027, and that money has to come from somewhere. Headcount is where it came from.

The Junior Developer Question

No part of the market has generated more anxiety than entry-level hiring. We covered this from two angles:

First, the 67% decline in junior developer openings between 2022 and 2026, based on Indeed and LinkedIn data. The article walks through the three forces driving the drop (AI code generation eating grunt work, companies over-hiring in 2021-2022, and the shift to hiring seniors who can supervise AI agents) and argues the panic is disproportionate to the actual displacement.

Then in June, we dug into what the 2026 employment data actually shows about AI replacement. The BLS numbers tell a more complicated story than the headlines: overall software developer employment is still growing, but the growth is concentrated in senior and AI-adjacent roles. The entry-level funnel has genuinely narrowed.

The through-line between these two pieces: juniors aren’t being replaced by AI. They’re being squeezed out by a market that now expects them to arrive with AI fluency, and the traditional “learn on the job” pipeline has constricted because companies are hiring fewer people into the roles where that learning used to happen.

This is a structural problem, not a cyclical one. In previous downturns (2001, 2008, 2020), junior hiring bounced back when the market recovered because companies still needed warm bodies writing CRUD endpoints and fixing bugs. In 2026, those tasks are the first to get automated. The entry-level work that used to serve as a training ground is disappearing even as the senior roles that the training was supposed to lead to are growing. The ladder still exists; the bottom rungs are being sawed off.

From where I sit in Cyprus, the European angle is particularly sharp. EU-based juniors face the same global compression, but with fewer of the advantages US juniors have: smaller VC-funded startup markets to absorb them, fewer internship pipelines at Big Tech (most EMEA internship programs were cut in the same layoff waves), and a credential culture that makes self-taught paths harder to navigate than in the US.

Salary Data: Cyprus, the EU, and the US

We published a dedicated salary analysis for Cyprus software engineers in 2026, covering what you’ll earn in Limassol and Nicosia across experience levels, the impact of the island’s growing fintech sector, and how Cyprus compares to other EU tech hubs. The short version: senior engineers in Limassol fintech companies are earning EUR 55-80K, competitive with Lisbon and Warsaw but below Berlin and Amsterdam. The non-dom tax regime still gives Cyprus an edge on effective take-home for EU-remote workers, though the government has signaled those benefits may narrow.

The broader SWE job market analysis covers the US and EU simultaneously, framing it as “two markets, one title.” The data shows a bifurcated market: AI-capable engineers are fielding multiple offers with 15-25% salary bumps, while traditional roles (frontend, backend without AI integration) are seeing flat or declining comp. The gap is wider in the US than in Europe, but the direction is the same on both sides of the Atlantic.

What’s unusual about the 2026 salary picture is that seniority premiums are growing even as total headcount shrinks. Companies cutting 10% of staff are simultaneously raising offers for the remaining hires. The skills premium for “can build and supervise AI agent workflows” has become the single largest comp differentiator in backend engineering, larger than the language premium (Go vs Python vs Java) or the domain premium (fintech vs enterprise) that dominated in 2024.

The Productivity Paradox

One of the most counterintuitive findings we covered: a METR study showing that AI coding tools make developers 24% faster in self-assessment but 19% slower in measured output. The gap between perceived and actual speed is the story. Developers genuinely believe they’re faster, and the tools feel productive in the moment, but the net output (measured by commit velocity, PR merge time, and feature completion) goes down.

We paired this with BCG’s research on AI “brain fry” across 1,488 workers: 47% of AI-heavy tool users reported cognitive fatigue that degraded their afternoon output. The productivity loss isn’t about the tools being bad. It’s about the cognitive overhead of context-switching between writing code and reviewing AI-generated code.

Together, these two pieces suggest that the productivity narrative around AI coding tools is more complicated than “AI makes you 2x faster.” For experienced developers on familiar codebases, the tools do speed things up. For complex, unfamiliar, or architecturally sensitive work, the overhead can eat the gains.

This has direct career implications. If companies are making headcount decisions based on the assumption that AI tools deliver a consistent 2x productivity boost, and the actual boost is closer to 0.8x for many workflows, the resulting understaffing creates burnout in the remaining team. Several of the layoff articles above note that post-layoff teams are being asked to do more with less, with AI tools cited as the reason they supposedly can. When those tools don’t deliver the expected output, the humans absorb the gap.

Reading Order

If you’re catching up, here’s a suggested path through the nine articles:

  1. Tech Layoffs Q1 2026 — the scale and shape of the cuts
  2. Big Tech Trading Jobs for AI Infrastructure — why the layoffs are capex-driven
  3. Meta Layoffs May 2026 — the biggest single-company cut
  4. Junior Developer Jobs in 2026 — the entry-level squeeze
  5. AI Replacing Junior Developers — what BLS data actually shows
  6. SWE Job Market 2026 — two markets, one title
  7. Cyprus Developer Salary 2026 — EU/Cyprus salary benchmarks
  8. AI Coding Productivity Paradox — 24% faster feel, 19% slower reality
  9. AI Brain Fry — cognitive fatigue from AI tools

What We Haven’t Covered Yet

Several threads in this cluster remain open:

  • Remote work policy shifts in 2026. RTO mandates are accelerating at the same companies doing layoffs. Is there a connection? The Glassdoor and Blind data exist but we haven’t pulled it together yet.
  • European AI Act impact on hiring. The Act’s compliance deadlines are now hitting, and companies are hiring for AI governance roles. Nobody has quantified how many new positions this creates versus how many it constrains.
  • Freelance and contract market for AI-displaced workers. Upwork and Toptal data suggest a surge in contract AI work, but the quality and comp data are murky.
  • Bootcamp and self-taught developer outcomes in the AI era. The pipeline data we’ve covered focuses on CS degree holders. The bootcamp cohort may be experiencing something different.
  • The senior engineer retention problem. If companies are cutting juniors and mid-levels while raising senior offers, senior engineers are picking up more coordination and mentoring load with less support. The burnout dynamics are different from what the BCG study measured (which focused on AI tool fatigue, not team structure fatigue).
  • Salary trajectories for AI-specialized roles versus traditional SWE. We have a snapshot of the gap in 2026, but the trajectory data (is the gap widening, stabilizing, or closing?) isn’t in our coverage yet.

Sources

Bottom Line

The nine articles in this cluster tell a consistent story: AI isn’t destroying tech careers wholesale, but it’s reshaping them faster than most people expected. The layoffs are real (180K+ in H1 2026), the junior pipeline is narrowing (67% fewer openings), and the productivity tools are more complicated than the marketing suggests (19% slower by one measure). At the same time, AI-fluent engineers are earning more than ever, and the companies doing the cutting are hiring aggressively for different roles.

If you’re a developer reading this in 2026, the strategic play hasn’t changed much from what it’s always been: get good at the thing that’s hardest to automate. Right now, that’s system design, cross-team coordination, and the ability to supervise AI agents effectively. The tools change; the principle doesn’t.

One pattern I keep seeing across these articles: the people who are thriving in the current market aren’t the ones who adopted AI tools earliest or most enthusiastically. They’re the ones who have deep enough domain knowledge to know when the AI output is wrong. That judgment, the ability to look at AI-generated code and say “this will break under load” or “this doesn’t handle the edge case where the user has two accounts,” is the actual skill moat. And it comes from years of writing and debugging code by hand, which is exactly the experience the shrinking junior pipeline threatens to cut off.

We’ll continue updating this cluster as the market evolves. The next pieces in the pipeline are remote work policy data for 2026 and the EU AI Act’s impact on tech hiring. Check back, or subscribe to the newsletter for weekly updates.