The Information Machine

Claude Code's product lead talks usage limits, transparency, and the "lean harness"

Ars Technica AI · Samuel Axon · 2026-05-15

Anthropic head of product Cat Wu explains why Claude Code has no long-term roadmap, defends the doubled usage limits announced at the Code with Claude conference, and describes the product's "lean harness" philosophy.

Open original ↗

Extraction

Topics: claude-codeai-developer-toolsanthropicproduct-strategy

Claims

  • Anthropic deliberately avoids a long-term roadmap for Claude Code, expecting that model capability improvements will outpace any fixed plan.
  • Anthropic doubled usage limits for Claude Code Pro and Max plan users in response to widespread user frustration over a compute crunch.
  • The second annual Code with Claude conference introduced new Managed Agents features and announced a compute deal with SpaceX.
  • Claude Code is built around a "lean harness" philosophy that prioritizes flexibility over prescriptive structure.

Key quotes

Anthropic doesn't have a long-term road map for Claude Code... it's betting that such a plan would be rendered moot by improvements in model capabilities and new signals from developers on how best to use it.