AI has become one of the most common explanations for tech job cuts, but the hiring picture for engineering roles is more complicated. New data cited by SignalFire suggests that software engineering, often described as one of the jobs most exposed to automation, has remained unusually resilient.
The finding does not settle the broader debate over AI and employment. It does, however, challenge the simple idea that AI coding tools are already translating into a broad collapse in engineering hiring.
Why the AI jobs debate is so heated
Whether AI is already replacing jobs remains a contested question. Tech layoffs reached their highest single month total in years in May, and AI was the most-cited reason, according to outplacement firm Challenger, Gray & Christmas.
Software engineering sits at the center of that debate because AI-powered coding tools have spread quickly. If a tool can help write code, review code, or generate software components faster, it is natural to ask whether companies will need fewer engineers.
But SignalFire’s researchers say the hiring data does not match the most extreme version of that fear. Asher Bantock, SignalFire’s head of research, described a mismatch between the public explanation for many layoffs and what the firm sees in workforce movement.
“The rationale given for lots of layoffs is consistently AI, and specifically they’ll say AI with respect to code; they’ll say one engineer could do the job of however many engineers in the past,” said Asher Bantock, SignalFire’s head of research. “What we’re seeing on the ground is a little inconsistent with that.”
What SignalFire measured
SignalFire’s analysis tracked the careers of millions of employees across more than 80 million companies. Instead of centering the analysis on layoffs, the firm focused on hiring data.
That choice matters because layoffs can be hard to measure in real time. People often delay updating their employment status after losing a job, so layoff data can lag the actual labor market. Hiring data can provide a clearer view of where companies are still adding workers.
According to SignalFire’s latest “State of Talent Report,” engineering was the most resilient job function in 2025. Total hiring across large tech companies dropped 25% compared to 2019 levels, but engineering roles declined by only 11%.
That smaller decline is important because it shows engineering did not fall as sharply as overall hiring. In a contraction, companies appear to be cutting back on hiring broadly while still prioritizing technical talent.
Engineers took a larger share of new hires
The trend is especially visible among the 12 companies SignalFire classifies as “Tech Majors”: Alphabet, Meta, Apple, Amazon, Microsoft, Netflix, NVIDIA, Tesla, Uber, Airbnb, Block, and Stripe.
Across those companies, engineers made up 55% of all new hires in 2025. In 2019, engineers represented 46% of new recruits, according to the report.
That shift means engineers accounted for a larger share of hiring even as total hiring fell. In other words, the hiring mix moved toward engineering rather than away from it.
SignalFire also found continued demand at early-stage startups. Those startups collectively hired 7% more engineers in 2025 than they did in 2019.
The startup data matters because early-stage companies are often under pressure to move quickly and spend carefully. If AI were already substituting broadly for engineering talent, early-stage startups would be one place to expect a clear pullback. SignalFire’s data points in the opposite direction.
What the data suggests about AI and engineering work
Bantock’s argument is straightforward: if AI were truly replacing engineering talent, engineering hiring should be among the first areas to weaken during the current tech hiring contraction. Instead, SignalFire’s data shows engineering headcount growing faster than most other job functions in tech.
This does not mean AI has no effect on software engineering jobs. The source data does not prove that every engineering role is secure, nor does it show that every company is hiring engineers at the same pace. It shows that, in the hiring patterns SignalFire analyzed, engineering has been more resilient than the broader tech labor market.
Other executives and researchers cited in the source also describe a more measured picture. Anthropic CEO Dario Amodei warned last year that AI could wipe out half of all entry-level white-collar jobs and push unemployment as high as 20% within five years. Yet Anthropic’s head of economics, Peter McCrory, told TechCrunch in March that he had not yet seen significant AI-driven effects on the workforce.
“There’s at least no larger material difference in unemployment rates” between workers who use Claude for the “most central task of their job in automated ways” — like technical writers, data entry clerks, and software engineers — and workers in jobs less exposed to AI that require “physical interaction and dexterity with the real world.”
Nvidia CEO Jensen Huang has rejected the idea that AI will replace engineers. In an interview at the Stanford Graduate School of Business in April, Huang said that all engineers at Nvidia are using agentic AI and that “software engineers are busier than ever,” according to the source.
Huang’s explanation is that agents can write code near instantaneously, but that this pushes engineers toward “the next idea.” In that framing, AI expands the pace and scope of software work rather than simply shrinking the need for people who do it.
The Jevons paradox lens
The source frames the current moment for engineering as an example of the Jevons paradox. That is the idea that greater efficiency can increase demand for a resource because more work becomes possible.
Applied to engineering, the logic is simple. If AI makes engineers more productive, companies may not stop needing engineers. They may instead ask engineers to build more, test more ideas, and move faster.
That view fits Bantock’s summary of the moment: “They’re suddenly a lot more productive, and there’s endless work for them to do.”
For now, the clearest conclusion is not that AI has no labor impact. It is that the strongest claims about AI already wiping out engineering jobs are not reflected in the SignalFire hiring data cited here. Engineering jobs remain exposed to AI, but they also appear to be among the most durable roles in tech hiring.