Anthropic Launches Claude Code Projects in Beta: Parallel Cloud Sessions That Keep Running After You Close …
What changed
Anthropic has revamped Projects in Claude Code, moving from a static folder model to an active conversation-driven system. Instead of managing a bundle of files with a single chat, users now have one ongoing conversation with Claude acting as a coordinator. When users describe work within this conversation, Claude splits tasks into separate threads. Each thread runs as its own Claude Code cloud session, continuing independently even after the user closes their laptop.
Why builders should care
This shift tackles a common pain point for developers and teams juggling multiple AI tasks concurrently. Older project formats forced workflows into a single thread, making it harder to switch between tasks or keep work alive beyond local interaction. Claude’s new approach creates parallel, persistent cloud sessions that keep code and conversations live and segmented. This supports better multitasking and asynchronous collaboration, especially for remote or distributed teams who need their projects to keep running without active supervision.
The practical takeaway
Users get a more flexible environment where AI-assisted coding and conversation threads run independently in the cloud. It decouples the work from the user’s immediate device state, reducing interruptions and manual context switching. Builders can spin off multiple ongoing Claude sessions for different tasks or features within a single project, streamlining how they organize, monitor, and update their work. This can lead to faster iteration cycles and fewer lost states due to device shutdown or network issues.
What to watch next
Monitor how Anthropic continues developing session management and collaboration features in Claude Code. Integration with existing IDEs or version control workflows will be key to adoption. Also watch for user feedback on thread coordination—the real test will be how well Claude manages task prioritization and context retention across multiple parallel conversations without causing confusion or performance drag. Finally, see if this approach pressures other AI coding platforms to improve their multitasking and session continuity.
AI Quick Briefs Editorial Desk