Skip to content
Agents tracked: 501 Releases (7d): 777 Agent status: no reported issuesSince last limit reset: Claude 10d · Codex 8d For AI agents: Markdown Updated Nov 21, 2026
Showcase AnthropicResults as reported by Anthropic

16 parallel Claude agents wrote a C compiler that builds the Linux kernel

An Anthropic researcher ran a team of Claude agents in a loop on a shared repository to write a Rust-based C compiler from scratch. The write-up and the code are both public.

The problem

It tests whether agents working mostly without supervision can finish a large, tightly specified engineering project when good tests and an oracle compiler keep them honest.

How they did it

  1. Several Claude instances each run in an infinite loop, picking the next task by themselves.
  2. Git and lock files keep agents from working on the same thing.
  3. Strong test suites, CI and GCC as a reference oracle tell the agents whether their output is correct.

Results as reported by Anthropic

  • Over nearly 2,000 Claude Code sessions and $20,000 in API costs, the agents produced a 100,000-line compiler.
  • It can build a bootable Linux 6.9 on x86, ARM, and RISC-V.
  • It reaches a 99% pass rate on most compiler test suites and also compiles QEMU, FFmpeg, SQLite, Postgres and Redis.
  • It is not a drop-in replacement for GCC. It relies on GCC for 16-bit x86 code and produces less efficient code.

These figures are as reported by Anthropic in Anthropic Engineering ↗ (Feb 5, 2026). Anthropic sells the product used here. AgentGid did not measure or independently verify them.

Takeaway. With a strong test harness, autonomous agent teams can take on projects of compiler size.
AgentGid's take

This suits a well-funded team with strong engineering skills that already has a mature test suite and a reference implementation to check against. The $20,000 in API costs is far beyond the $20/mo Claude Code plan price, so budget for pay-as-you-go usage. Without an oracle like GCC, the agents have nothing to keep them honest. For a cheaper way to try the loop on a smaller project, Pi (Free (OSS)) charges only model provider costs.

The agent used here

Similar use cases

Guides