The Developer Shortage Is Real. The Developer You Need Has Changed.
The shortage is not of developers in general. It is of a specific kind of developer, and that distinction is reshaping hiring.
Two things are true about software developer demand right now, and most conversations only pick one of them.
The first is that overall demand keeps growing. The second is that the profile of who gets hired has shifted hard enough that treating this as one continuous market, the same as five years ago but slightly bigger, misses what is actually happening. The shortage is real. It is not a shortage of developers in general. It is a shortage of a specific kind of developer, and the gap between the two is where most hiring decisions are currently going wrong.
What the growth numbers actually say
The headline figures are genuinely strong. The U.S. Bureau of Labor Statistics has projected software developer employment growth in a range depending on the specific occupational category and time period measured, with recent projections in the area of 15 percent growth from 2024 to 2034, a figure well above average growth across all occupations. The global software market itself is on a similar trajectory, projected to approach 900 billion US dollars by 2030.
None of this suggests a market in decline. What it does not tell you, on its own, is where that growth is concentrated, and the answer to that question matters more than the headline number.
Where the growth is actually happening
AI and machine learning engineering roles are growing at a rate that dwarfs the rest of the market. Industry hiring data has tracked triple digit year over year growth in AI and ML job postings over the past two years, with agentic AI roles specifically growing even faster than the broader AI and ML category. Demand for backend and infrastructure engineering remains strong and represents the largest hiring category by raw volume.
Generalist full stack roles, by contrast, show softer demand growth than the categories above them, a pattern that industry analysis attributes largely to AI coding tools extending how much a single developer can cover. The effect is not that full stack developers are less valuable. It is that AI assistance narrows the gap between what one strong developer can deliver and what previously required a larger team, which changes the shape of hiring even where overall demand is healthy.
“The shortage is real. It is not a shortage of developers in general. It is a shortage of a specific kind of developer.”
The sharper story at entry level
The most disputed area of this data, and the one worth treating with the most care, concerns junior and entry level hiring.
The estimates vary enormously depending on the source, and the variation itself is worth understanding before drawing conclusions. A study out of Harvard, examining generative AI adoption across a very large sample of workers, found a meaningful but moderate effect: junior developer employment declining by roughly 9 to 10 percent within six quarters of AI adoption at a company. Stanford's AI Index research points to a sharper effect specifically among the youngest cohort of developers, with employment for developers aged 22 to 25 down by close to 20 percent since 2022. A number of industry commentary pieces cite far larger figures, in some cases suggesting entry level hiring has collapsed by well over half, but these figures come from less rigorous sources and should not be treated with the same confidence as the university research above.
What is consistent across every credible source, regardless of the exact magnitude, is the direction and the mechanism. The bar for entry level hiring has risen. Companies increasingly expect junior candidates to arrive with a level of systems understanding and AI fluency that previously would have been learned on the job over the first year or two of employment. Some organisations have responded by cutting junior hiring substantially. Others, including at least one major technology company that has publicly committed to expanding entry level hiring, are betting that the companies which maintain their junior pipeline now will have an advantage when the resulting shortage of mid level talent arrives in three to five years, as several industry analysts expect it will.
What this means for how hiring decisions should actually be made
Put together, the data points toward a specific and fairly clear conclusion. Total demand for developers is not shrinking. The definition of what makes a developer worth hiring is narrowing around a specific combination: genuine engineering judgement, paired with real fluency in working alongside AI tooling, rather than either skill on its own.
A developer with strong traditional engineering skills but no meaningful AI Native habits is competing in a market that increasingly expects both. A developer with AI fluency but without the underlying engineering judgement to direct it well is, if anything, in a worse position, since AI assistance amplifies existing skill rather than replacing the need for it. The developers seeing the strongest demand are the ones who combine both, and that combination does not reliably correlate with years of experience, which is a separate argument worth making on its own.
This has a direct implication for how companies approach contractor hiring specifically. The instinct to treat contractor selection as primarily a cost decision, comparing day rates and filling a headcount gap as cheaply as possible, is increasingly out of step with where the actual value sits in this market. The developers worth paying for are the ones who reduce the overhead a placement generates, not the ones who simply cost the least per day, and the research on where genuine demand and genuine scarcity sit in the market reinforces exactly that point rather than contradicting it.
The honest caveat
Forecasts about a market this closely tied to the pace of AI development carry real uncertainty, and the wide range in the entry level hiring figures above is itself evidence of how much disagreement exists even about data describing what has already happened. The direction of the trend is consistent across credible sources. The exact magnitude is genuinely unsettled, and any claim stated with more precision than the underlying research supports should be treated with appropriate scepticism.
What is not in serious dispute is the shape of where demand is heading. Growth is real. It is concentrated around a specific combination of skills rather than spread evenly across the market. And the companies that adjust their hiring approach to reflect that combination, rather than continuing to hire the way the market worked five years ago, are the ones positioned to benefit from the growth that is genuinely there.
“The developers worth paying for are the ones who reduce the overhead a placement generates, not the ones who simply cost the least per day.”
At Navigaite, this is the exact combination we assess for in every placement: genuine engineering judgement paired with real AI Native fluency, not one without the other. If your team is trying to hire into a market that has shifted this much, the assessment process matters as much as the search itself.
Read next
Why the Cheapest Contractor Is Almost Always the Most Expensive Hire
The day rate is almost never the number that determines whether a contractor engagement was worth it.
Field ReportThe Toolchain She Had Never Used, on Day One
The difference between tool fluency and AI-native judgement, seen in the field.
Thought LeadershipWhat AI-Native Systems Actually Look Like in Production
The structural differences between teams that built AI into how they deliver and those that bolted it on.
Hiring into a market that has shifted?
Navigaite assesses genuine engineering judgement and real AI Native fluency together, because one without the other is no longer enough.
Get in touch
