Employment for software developers in their early 20s has dropped sharply since generative AI tools flooded the market. Older engineers keep advancing. The split shows up clearly in fresh data.
By July 2025, jobs for those aged 22 to 25 in software development had fallen nearly 20% from late 2022 peaks, according to a Stanford Digital Economy Lab study. For high AI-exposure roles overall, employment declined 6% for that young group. It rose 9% for workers aged 35 to 49. The pattern holds.
Data Reveals a Clear Generational Divide
More recent figures sharpen the picture. ADP and Stanford researchers found employment in the most AI-vulnerable occupations contracted 4.2% for ages 22 to 25 as of April, compared with a 1.7% drop for similar-aged workers in less exposed fields. Software development stands out. Young roles tumbled after ChatGPT’s arrival. Positions for those over 30 continued climbing. (Yahoo Finance, June 2026)
But total software developer jobs aren’t vanishing. Postings on Indeed rose 11% annually, outpacing overall growth. The Bureau of Labor Statistics projects 15% expansion by 2034. Five times the average rate. Companies talk of producing more software than ever because AI lets nearly anyone generate code. That demands seasoned engineers to direct, integrate and refine it. (CNN, April 2026)
IBM is tripling its U.S. entry-level hiring, including for software developers. “Junior engineers are now capable with AI of taking on tasks that once required experienced developers,” said Neel Sundaresan, IBM’s general manager of automation and AI. Intuit’s chief technology officer Alex Balazs noted the company seeks early-career developers who grew up with these tools. They may grasp them better than mid-career staff.
Still, stories circulate of new graduates sending thousands of applications with no offers. One 2023 computer science major applied to 5,762 tech jobs without landing a full-time role. The New York Times captured the frustration in 2025. Bootcamp grads who walked into junior positions in 2023 now compete against AI that handles routine coding faster and cheaper.
Productivity gains explain part of the shift. AI coding assistants manage basic tasks, debugging and test writing. Seniors handle more with their help. One experienced developer now completes work once done by several juniors. Teams review AI output, but that review still needs judgment. Juniors often lack the experience to spot flaws or recover from errors.
And here’s the tension. Four in 10 developers reported in 2025 that AI had already expanded their opportunities. Nearly seven in 10 expect their roles to evolve further in 2026. They move from routine coding toward architecture, system integration and higher-level decisions. (World Economic Forum, January 2026)
Yet entry-level openings now represent about 20% of software positions, down from 30% in 2023. Big Tech cut fresh graduate hiring by roughly 50% over three years. Unemployment for recent computer engineering grads hit 7.5%, nearly double the national rate. Harvard research on 62 million workers showed companies adopting generative AI reduced junior developer hiring by 9% to 10% within six quarters.
Critics point out other factors. Layoffs, offshoring to lower-cost regions and post-pandemic normalization play roles. Some analysts argue the junior pipeline matters more than ever. AI generates more code. That code requires maintenance, integration and understanding when it breaks. Remove the entry layer and seniors become expensive janitors. Future technical leaders disappear.
Skills Gap Widens as Expectations Shift
Surviving juniors must demonstrate AI fluency from day one. They pair tools with critical thinking. Blind trust in outputs fails. Teams that train young developers to question AI, experiment locally and own workflows gain advantages. Local-first open-source stacks avoid token costs and keep code private.
McKinsey’s 2025 survey found a third of organizations expect AI to shrink their workforce in the coming year. Yet BLS forecasts strong long-term demand. The resolution sits in transformation, not elimination. Engineers become AI-augmented system designers. Gartner predicts 80% of the engineering workforce must upskill by 2027.
Experience still counts. Judgment separates useful AI suggestions from costly mistakes. Young developers who master this stand out in a crowded field. Those who don’t face longer job searches and lower starting expectations. Median offers may settle around $80,000 in some markets, far from the $165,000 tech salaries once chased.
Companies experiment. Some replace routine junior tasks with AI agents. Others hire AI-native talent who accelerate faster. The data shows a fork in the road. One path leads to fewer entry points and steeper ramps. The other invests in training the next cohort to direct increasingly capable systems.
Software development led this change. Knowledge work follows. What happens to young coders today signals broader patterns for analysts, writers, designers. Productivity rises. Entry barriers shift. The ladder doesn’t disappear. It gets harder to reach the first rung.
Recent discussions on X echo the divide. Engineers debate whether cutting juniors saves costs or stores up technical debt. One post noted AI writes code but creates more to review. Another warned that without juniors, systems grow opaque even to seniors. Real outcomes will show in the next hiring cycle.
WebProNews is an iEntry Publication