Anthropic study finds junior technical hiring fell sharply
Anthropic's internal data shows junior roles fell from 27% to 9% of its technical hires as AI agent hours grew to six times employee hours, an early look at how AI is changing work inside a frontier lab.
Anthropic has published research drawn from its own workforce. It reports that the company has cut back on junior technical hiring while AI agents take on more of its work. The report's authors are Nathan Wilmers, Regina Catipon, Maxim Massenkoff and Peter McCrory. It draws on internal data covering employment, AI use, output metrics, surveys, calendars and hiring records, along with 17 employee interviews.
Junior employees fell from 21% to 14% of staff in Anthropic's technical departments after early 2025. Junior roles fell faster as a share of technical hires, from 27% to 9%. Postings for junior roles have dropped sharply since December 2025. By April, engineering and research teams had deprioritised lower-band technical roles. In a survey that month, about a third of ML managers and a fifth of SWE managers said they were hiring less because of AI.
The report says interactive agent hours were six times employee hours by August. Output per person rose for code, closed sales and processed contracts. When the same employees were tracked over time, their output went up after they adopted new AI tools. The authors say human attention is now the bottleneck: reading everything their Claude agents produce would take non-technical staff half their work week. Coding-focused agent sessions among non-technical employees grew from 2 a month in August 2025 to 90 in May 2026.
Hiring managers now put more weight on autonomy and judgment than on technical expertise. Anthropic says it has set up an early warning system to track jobs at risk and is prioritising internal redeployment.
The findings cover a single company and come from Anthropic's own analysis. The adoption comparisons are observational, not randomised.