Anthropic is hiring fewer junior engineers, and its staff now spend more time in meetings than they did a year ago. TechCrunch AI reports that the company published an internal workforce report on Friday. Anthropic says the report could preview “how AI will shift work organization, hiring, and labor demand across the economy.”
Read that claim with some caution. Anthropic sells Claude, so it gains when people believe AI changes how work gets done. Still, few companies have “thoroughly molded its workflows around AI” the way Anthropic says it has. That makes the numbers worth a close look.
📉 The junior pipeline is shrinking
The headline numbers are about who gets hired. According to TechCrunch AI:
- Junior staff in technical departments fell from 21% to 14% since early 2025.
- Junior technical hires dropped harder, from 27% to just 9%.
- Technical teams as a share of the workforce slipped from 61% to 56%.
Anthropic gives two reasons for that last shift. Business functions grow naturally as a company scales. And coding tools now let non-technical employees “make software themselves.”
What stands out is the hiring number. A drop from 27% to 9% isn’t a small adjustment. It means the company has mostly stopped bringing in junior engineers. If other AI-heavy companies follow, the usual entry path into software careers gets a lot narrower.
🍼 From doing work to “babysitting” Claudes
The survey data shows how the work itself feels different. In an April 2026 survey, 27% of employees said Claude made them more productive. And 20% said their job had shifted “from actually doing work to managing Claudes.”
One engineer compared it to babysitting: “When you are babysitting, every child brings to you concerns with an equal demand for your attention and the various [Claudes] also do that.” Another described an “eternal pressure at Anthropic to move as fast as possible and use Claude for everything.”
This is significant because it moves the bottleneck. The report says “human attention and judgement” now limit productivity, because Claude produces a “deluge” of text that someone has to read and sort. Producing output has become cheap. Checking it hasn’t.
🤝 Why meetings went up, not down
Many people expected AI to cut coordination overhead. Anthropic’s calendar data shows the opposite:
- Meeting time per employee is up 20% year over year.
- For technical contributors, it’s up 16%.
The company puts this down to “human trust and accountability.” Employees kept calling “face-to-face coordination” indispensable. When agents do more of the execution, people spend more time deciding what to build, who owns it, and whether the results can be trusted.
Anthropic also says productivity went up, with more code per person, more sales closed per rep, and more contracts processed. To its credit, the company admits that “more code does not necessarily mean better code or business impact,” though it says it hasn’t seen signs that quality is slipping.
🔮 What the next 1-3 years likely look like
If Anthropic is an early indicator, here’s what many teams can expect:
- Entry-level roles change shape. Fewer juniors get hired to write boilerplate code. Those who do get hired will be judged on review skills and judgment, not on how much code they produce.
- Review becomes the scarce skill. Teams that can’t absorb AI output quickly will fall behind, however many agents they run.
- The line between technical and non-technical roles blurs. Sales, legal, and ops staff will build their own tools.
- Coordination gets more valuable. Expect more meetings, not fewer, until verification tools catch up.
✅ What to do now
- Measure review load, not just output. Track how much time your people spend reading and checking AI work.
- Don’t kill your junior pipeline without a plan. Senior engineers have to come from somewhere. Build training paths around agent supervision.
- Set up triage. Give “Claude managers” ways to prioritize, so every agent isn’t asking for equal attention.
- Watch quality metrics closely. “More code” is easy to measure, and that’s exactly why it can mislead you.
Anthropic’s report is one company describing itself. But it’s one of the clearest early looks at an AI-first organization from the inside. The full details are available in TechCrunch AI’s original coverage.