Anthropic open-sourced the shared-memory design behind its internal 30,000-agent fleet, where every…
anthropic open-sourced the shared-memory design behind its internal 30,000-agent fleet, where every branch thread syncs specs, decisions, and team preferences in real time. parallelism without shared state was the wall, and this is the fix.
the agent that remembers becomes the one you trust.
Context
Anthropic's Claude Code projects post says every thread now adds to and draws from a shared memory, that threads are full Claude Code sessions, and that a project library collects files and artifacts. Anthropic's pace-of-AI-development post says about 30,000 agents were doing research and engineering work at once on its most-used internal platform as of August 2026.
Shared project memory in the product is first-party. That Anthropic open-sourced the design was not found: no Anthropic repository, spec or post announcing it was located, and the only hits were third-party snippets. The 30,000-agent platform figure and the product's shared memory are two separate statements, and nothing was found connecting the product's memory to the internal fleet's design. Syncing specs, decisions and team preferences in real time is not in the text read, which says memory, decisions and results build up in the project.
Related work
Watch next
- An Anthropic engineering post or licensed repository for the memory design.
Sources
- Projects, redesigned (Anthropic)claude.com
- Measuring the pace of AI development (Anthropic)anthropic.com
Provenance
The note above is reproduced unedited from the original post, first published on Threads on 18 September 2026 at 08:35 IST. Sources are the papers and datasets the note draws on.
View the original post ↗Embed this note
More notes
The air is now being asked to keep its own ledger
the air is now being asked to keep its own ledger: ecmwf’s aifs compo becomes the first ai model to forecast atmospheric composition globally every three hours, cleanair simulates 365 days of pm2.5 over china in ten seconds, and a unified framework maps six pollutants at one kilometer across the whole country. the air now files its own composition report.
read the note →The current is now being asked to draw its own map
the current is now being asked to draw its own map: china’s langya 2.0 predicts six ocean phenomena including internal waves and mesoscale eddies, a deep net called wenhai resolves eddies globally with air sea flux formulas built in, and scripps infers surface currents from the way temperature patterns deform in satellite images. the ocean now files its own circulation report.
read the note →The soil is now being asked to report its own carbon
the soil is now being asked to report its own carbon: a nix color sensor paired with generative data augmentation predicts soil organic carbon without a lab, random forest drives 74 percent of soil health mapping studies, and sentinel 2 tracks five year carbon change across france and italy from 922 samples. the dirt now files its own carbon account.
read the note →