MeshXFS is a clustered open-source file-system in which "no human wrote any of its own code". Depending upon your views of AI, not necessarily something you want to hear for entrusting your data in an AI-created solution. Instead, Claude Opus was tasked with adapting the XFS file-system driver into a clustered file-system.
Steve Shoecraft who started the MeshXFS project and wrote into Phoronix to highlight the initiative. He remarked:
"MeshXFS began as an experiment to answer one question: can AI write a kernel filesystem driver, and not a simple one, but a clustered filesystem with distributed locking, concurrent access from many nodes, fencing and crash recovery? It is now a working, released, open-source (GPL-2.0) POSIX clustered filesystem, and the answer so far is yes."
XFS was adapted into a POSIX clustered file-system complete with distributed locking, many node concurrent access capabilities, fencing and crash recovery, and other features. Steve further explained of this MeshXFS creation:
"It starts from the upstream XFS code, written by the XFS developers. Everything that turns XFS into a filesystem many machines can mount read-write at once was written by AI:- two distributed lock managers: one over the network, and one that keeps lock state on the shared disk itself using SCSI COMPARE AND WRITE, with no lock network at all
- cache coherency across nodes, fencing with SCSI persistent reservations, and survivor-led journal recovery
- an on-disk envelope around XFS, plus the userspace tools (mkfs, fsck, resize)
- every change to the forked XFS code since the fork (the xfs/ tree has grown from about 226k to about 314k lines)
- the test harnesses, the test rig automation, and the release validationClaude Opus, running as an autonomous coding agent in Claude Code, did a large share of the work, consulting Claude Fable, GPT, Gemini and Astra on the hardest problems: lock protocol design, recovery races, fencing correctness. The AI also wrote and ran the tests, diagnosed the failures and fixed them. My part was setting the goal and the bar: data integrity first, never crash the host, and stay within twice the time native XFS takes on the same storage. I did not write, edit or review the code."
For those concerned about entrusting their data into an AI-modified file-system, each MeshXFS release is said to undergo a ~30 row suite on real multi-node clusters through a variety of different real-world scenarios.
Those curious about MeshXFS and wanting to learn more can do so via the project's GitHub.