SignalFire data shows engineering hiring is more resilient than almost anyone expected

A new SignalFire report analyzing hiring data across 80 million companies finds that engineering roles have declined only 11% from pre-pandemic levels, compared to a 25% drop in total tech hiring. The report argues that AI tools are making engineers more productive, leading companies to hire more engineers rather than replace them, with engineers now comprising 55% of new hires at major tech firms, up from 46% in 2019.

A new SignalFire report tracking 80 million companies finds engineering roles have weathered the AI era far better than total tech hiring, and the most AI-forward firms are actually adding engineers, not cutting them. The narrative has been loud and persistent: AI is coming for software engineers. Developers will be automated away. The next generation of startups will run on a skeleton crew and a stack of large language models. SignalFire's 2026 State of Talent Report, published this week and drawing on hiring data across more than 80 million companies, says that story is mostly wrong, and the firms betting on it may be making a serious strategic error. Total tech hiring across what SignalFire calls the "Tech Majors" including Alphabet, Meta, Nvidia, and Stripe has fallen to 75% of its pre-pandemic baseline, a 25% drop from 2019. Engineering, by contrast, declined only 11%. That gap is not noise. It reflects something structural about how AI actually interacts with technical work, and SignalFire's head of research Asher Bantock frames it in terms of a concept economists have been arguing about since the 1860s: the Jevons Paradox. William Stanley Jevons noticed, back when Britain was worrying about coal, that making steam engines more efficient did not reduce coal consumption. It increased it, because cheaper energy made more applications economically viable. Bantock's argument is that the same dynamic is playing out now with engineering talent. Make engineers more productive with AI tools and you don't eliminate positions, you expand the frontier of what's worth building. "They're suddenly a lot more productive," he noted of engineers in the current moment, "and there's endless work for them to do." The data backs that reading more concretely than most takes on this subject manage to do. Engineers now make up 55% of new hires at the Tech Majors, up from 46% in 2019. That is a meaningful compositional shift, not a rounding error. The companies most aggressively deploying AI tools are not replacing their engineering orgs; they're concentrating them. What's getting cut is everything around the edges: designers, marketers, generalist business roles. The technical core is holding, and in some cases growing. The early-stage data is where the report gets genuinely interesting for founders. You might expect young companies to be the fastest adopters of "AI-first, headcount-last" thinking, running lean on the assumption that a handful of engineers plus the right models can do what a full team once required. The numbers don't support that. Early-stage startups collectively hired 7% more engineers in 2025 than they did in 2019. They are also running flatter: the average engineering span of control has risen by 34%, meaning roughly 15 engineers per manager where there used to be closer to 11. But they are not shrinking their engineering headcount. They are stretching each engineer further and then hiring more of them anyway. That pattern should reframe how founders think about the "AI replaces developers" pitch. The instinct to hoard AI tools as a headcount substitute may feel like capital efficiency. What the data suggests instead is that the founders hiring engineers aggressively, and equipping them well, are the ones building faster, not the ones waiting to see if the tools get good enough to skip the people. There is a real caveat buried in the same report, though. Entry-level hiring has been genuinely hit. According to SignalFire's findings, new role starts by people with less than one year of post-graduate experience dropped 50% between 2019 and 2024. That is a significant structural shift, and it deserves to be taken seriously. AI has compressed the demand for early-career work: the tasks that once served as on-ramps, the glue code, the boilerplate, the mechanical debugging, are increasingly handled by tools. What remains scarcer is the judgment, context, and design sense that takes years to develop. The industry is not firing engineers. It is becoming less willing to pay to train them. For anyone running a startup right now, the practical implication is pointed. If you've been treating engineering headcount as the thing to minimize while AI tools handle the rest, TechCrunch's coverage of the SignalFire findings published today offers a useful corrective: the most AI-forward firms in the dataset are doing the opposite. They are betting that more capable engineers, working with better tools, can expand into larger and more ambitious technical problems. Bantock's Jevons framing is not just a clever analogy. It describes a real mechanism with real hiring consequences, and the data now spans enough companies and enough time to treat it as something more than a hypothesis. The engineering job market is not what the doomier headlines promised. It's something more complicated and, for experienced engineers at least, considerably more interesting. Also read: Qualcomm bets its future on a 250-core data center chip and a $3.9 billion software acquisition https://startupfortune.com/qualcomm-bets-its-future-on-a-250-core-data-center-chip-and-a-39-billion-software-acquisition/ • Micron's $41 billion quarter confirms the AI memory supercycle is nowhere near done https://startupfortune.com/microns-41-billion-quarter-confirms-the-ai-memory-supercycle-is-nowhere-near-done/ • Anthropic's distillation problem reveals that export controls alone cannot hold the line in the US-China AI race https://startupfortune.com/anthropics-distillation-problem-reveals-that-export-controls-alone-cannot-hold-the-line-in-the-us-china-ai-race/