Junior Share of Anthropic's New Technical Hires Falls From 27% to 9%

Anthropic published a report on October 16 (local time) analyzing how the makeup of its own workforce and the nature of its employees’ work have been changing. Among new technical hires, the share of junior staff has fallen, while the time employees spend in meetings has risen. The company drew on personnel records, Claude usage logs, work output, surveys, calendars, and recruiting materials. It also interviewed 17 employees.

How much has junior hiring dropped?

Since the start of 2025, the share of junior staff within technical departments has fallen from 21% to 14%. Among new technical hires over the same period, the junior share dropped from 27% to 9%. The shift was larger among incoming hires than in the existing workforce composition. The two declines amount to 7 percentage points and 18 percentage points, respectively.

Technical departments also account for a smaller slice of the overall headcount, down from 61% to 56% over the past year. The absolute number of technical hires, however, increased. Even as the company brought on more technical staff, both the technical share of the organization and the junior share of new hires declined.

If AI writes the code, what do people do?

Anthropic said it is handing off certain tasks, such as writing code and finding bugs, to AI agents that carry out work on their own. Employees instead review what the AI produces and manage the work. The company also found that staff spend more time on judgment calls and coordination. In August 2026, agents logged six times as many working hours as employees did.

Time spent in meetings is up as well. Calendar data shows that for the median employee, meeting time rose 20% compared with a year earlier. For individual contributors in technical roles, meeting time rose 16%. The company pointed to its rapid growth and an increase in meeting-heavy roles as contributing factors.

Anthropic also examined how AI was reshaping internal work in 2025. This time it widened the scope to hiring and organizational change by adding data on hiring composition, candidate interview assessments, agent working hours, and meeting time. The researchers noted that AI adoption was not randomized and coincided with a period of fast company growth. As a result, they said, the productivity gains cannot be attributed entirely to AI.

What happens to entry-level opportunities?

Anthropic said it tracks roles most exposed to change and prioritizes internal mobility for current employees. At the same time, it acknowledged that such measures alone will not resolve the slowdown in junior hiring. Helping existing staff move into new kinds of work and expanding entry points for people early in their careers, the company said, are two separate problems that need separate solutions.

In this analysis, employees’ roles shifted away from doing the work directly and toward reviewing output and coordinating. The open question is what that shift means for junior workers’ chances of getting hired and growing. It is a case study in what role is left for humans as AI takes on more of the work.