Per-developer environments were the goal. Agents moved the goalposts.

6.2 relevance

Directly addresses agent-driven development environments, a key interest.

AI/ML thenewstack.io

Answer the board on AI readiness

Arrive with evidence, not opinion. See how your foundation benchmarks against your peers, get a clear read, and leave with a ranked place to start.

Why It Matters

For platform engineers building developer environments, this shifts capacity planning from seat-based to change-based pricing, requiring infrastructure that can dynamically provision per-change tenants rather than per-developer namespaces.

Key Takeaways

Redesign your platform's tenant model around changes, not developers or agents, to avoid mispricing capacity and serializing concurrent workstreams.

Per-developer environments were the goal. Agents moved the goalposts.

Summary

Multi-tenancy has shrunk from organizations to teams to individual developers over 60 years, but coding agents break the person-tenant assumption by running multiple concurrent workstreams per developer. Anthropic ran nearly 2,000 Claude sessions across two weeks building a C compiler, and Cursor encourages running as many agents as desired in parallel. The new tenant is no longer the developer or agent but the individual change, as each concurrent workstream needs its own isolated environment with data, topics, and service versions.

Author

Arjun Iyer