GitLab Announces Project Switch for AI Agent Era
As AI agents increasingly perform parallel coding at massive scale, existing source code management tools face unprecedented load; Project Switch is GitLab's architectural response to that shift.
Reporting from 1 sources: GameBusiness.jp.
GitLab announced 'Project Switch,' a next-generation Git-compatible source code management service designed for AI agents, at the GitLab Transcend event in London on June 10. The service adopts a new architecture with compute-storage separation, achieving up to 50x faster performance and halving token consumption by reducing network traffic to 1/1000th.
GitLab unveiled Project Switch at its GitLab Transcend event in London on June 10, targeting the growing load that AI agents place on source code management services. Traditional Git services assumed human-paced manual saves and updates, but parallel AI agents now copy, generate, and save code far faster than manual work, causing slowdowns and instability. Project Switch keeps the Git protocol intact but replaces the backend with a compute-storage separated architecture, cutting network traffic to 1/1000th and halving token consumption. GitLab claims clone speeds improve up to 42x, write speeds up to 17x, and overall AI agent work time speeds up by roughly 22x. An intelligence layer routes requests, caches data, and handles background processing like bitmap updates to maintain stability under heavy loads.
Synthesized by Yomimono from the 1 cited source below, including Japanese-language reporting where cited, then editorially reviewed before publishing.