Anthropic Launches Claude Projects Beta With Parallel Code Sessions for Pro and Max Users
Updated
Updated · The New Stack · Sep 17
Anthropic Launches Claude Projects Beta With Parallel Code Sessions for Pro and Max Users
3 articles · Updated · The New Stack · Sep 17
Summary
Thursday’s beta gives select Claude Pro and Max subscribers a redesigned Projects tool that can turn one engineering goal into multiple Claude Code cloud sessions running in parallel.
The coordinator splits work, assigns separate branches and repositories to each thread, reviews outputs, and assembles the result, reducing the manual session management developers previously had to do themselves.
Each parallel thread counts as a full Claude Code session, so usage limits can be hit faster — a sensitive point after a class-action suit last week challenged undisclosed weekly ceilings on Anthropic’s Max plan.
Shared memory now carries project context across threads and over time, while users can track Project-specific usage and choose models and effort levels for both coordinator and worker sessions.
Anthropic plans to widen access over the coming week, add local-thread and mobile support soon, and later bring the feature to Team and Enterprise customers as it folds more tools into a single Claude interface.
Will Anthropic’s chat-first workspace finally dethrone Google and Microsoft, or is an all-in-one interface too chaotic for complex document creation?
As Claude’s Cowork processes tasks offline in the background, are we witnessing the quiet shift from AI assistants to fully autonomous digital employees?
With Claude now connecting directly to private enterprise data, can its new security sandboxes truly prevent catastrophic prompt-injection attacks?