{"id":413418,"date":"2026-02-07T19:51:10","date_gmt":"2026-02-07T19:51:10","guid":{"rendered":"https:\/\/www.newsbeep.com\/uk\/413418\/"},"modified":"2026-02-07T19:51:10","modified_gmt":"2026-02-07T19:51:10","slug":"sixteen-claude-ai-agents-working-together-created-a-new-c-compiler","status":"publish","type":"post","link":"https:\/\/www.newsbeep.com\/uk\/413418\/","title":{"rendered":"Sixteen Claude AI agents working together created a new C compiler"},"content":{"rendered":"<p>Amid a <a href=\"https:\/\/arstechnica.com\/information-technology\/2026\/02\/ai-companies-want-you-to-stop-chatting-with-bots-and-start-managing-them\/\" rel=\"nofollow noopener\" target=\"_blank\">push toward AI agents<\/a>, with both Anthropic and OpenAI shipping multi-agent tools this week, Anthropic is more than ready to show off some of its more daring AI coding experiments. But as usual with claims of AI-related achievement, you\u2019ll find some key caveats ahead.<\/p>\n<p>On Thursday, Anthropic researcher Nicholas Carlini <a href=\"https:\/\/www.anthropic.com\/engineering\/building-c-compiler\" rel=\"nofollow noopener\" target=\"_blank\">published a blog post<\/a> describing how he set 16 instances of the company\u2019s Claude Opus 4.6 AI model loose on a shared codebase with minimal supervision, tasking them with building a C compiler from scratch.<\/p>\n<p>Over two weeks and nearly 2,000 Claude Code sessions costing about $20,000 in API fees, the AI model agents reportedly produced a 100,000-line Rust-based compiler capable of building a bootable Linux 6.9 kernel on x86, ARM, and RISC-V architectures.<\/p>\n<p>Carlini, a research scientist on Anthropic\u2019s Safeguards team who previously spent seven years at Google Brain and DeepMind, used a new feature launched with Claude Opus 4.6 called \u201c<a href=\"https:\/\/code.claude.com\/docs\/en\/agent-teams\" rel=\"nofollow noopener\" target=\"_blank\">agent teams<\/a>.\u201d In practice, each Claude instance ran inside its own Docker container, cloning a shared Git repository, claiming tasks by writing lock files, then pushing completed code back upstream. No orchestration agent directed traffic. Each instance independently identified whatever problem seemed most obvious to work on next and started solving it. When merge conflicts arose, the AI model instances resolved them on their own.<\/p>\n<\/p>\n<p>The resulting compiler, which Anthropic has <a href=\"https:\/\/github.com\/anthropics\/claudes-c-compiler\" rel=\"nofollow noopener\" target=\"_blank\">released on GitHub<\/a>, can compile a range of major open source projects, including PostgreSQL, SQLite, Redis, FFmpeg, and QEMU. It achieved a 99 percent pass rate on the GCC torture test suite and, in what Carlini called \u201cthe developer\u2019s ultimate litmus test,\u201d compiled and ran Doom.<\/p>\n<p>It\u2019s worth noting that a C compiler is a near-ideal task for semi-autonomous AI model coding: The specification is decades old and well-defined, comprehensive test suites already exist, and there\u2019s a known-good reference compiler to check against. Most real-world software projects have none of these advantages. The hard part of most development isn\u2019t writing code that passes tests; it\u2019s figuring out what the tests should be in the first place.<\/p>\n","protected":false},"excerpt":{"rendered":"Amid a push toward AI agents, with both Anthropic and OpenAI shipping multi-agent tools this week, Anthropic is&hellip;\n","protected":false},"author":2,"featured_media":392269,"comment_status":"","ping_status":"","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[20],"tags":[554,733,4308,86,56,54,55],"class_list":{"0":"post-413418","1":"post","2":"type-post","3":"status-publish","4":"format-standard","5":"has-post-thumbnail","7":"category-artificial-intelligence","8":"tag-ai","9":"tag-artificial-intelligence","10":"tag-artificialintelligence","11":"tag-technology","12":"tag-uk","13":"tag-united-kingdom","14":"tag-unitedkingdom"},"_links":{"self":[{"href":"https:\/\/www.newsbeep.com\/uk\/wp-json\/wp\/v2\/posts\/413418","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.newsbeep.com\/uk\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.newsbeep.com\/uk\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/uk\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/uk\/wp-json\/wp\/v2\/comments?post=413418"}],"version-history":[{"count":0,"href":"https:\/\/www.newsbeep.com\/uk\/wp-json\/wp\/v2\/posts\/413418\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/uk\/wp-json\/wp\/v2\/media\/392269"}],"wp:attachment":[{"href":"https:\/\/www.newsbeep.com\/uk\/wp-json\/wp\/v2\/media?parent=413418"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.newsbeep.com\/uk\/wp-json\/wp\/v2\/categories?post=413418"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.newsbeep.com\/uk\/wp-json\/wp\/v2\/tags?post=413418"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}