According to Anthropic (@AnthropicAI), Opus 4.6 was tasked with using agent teams to autonomously build a C compiler, with minimal human oversight. After two weeks, the compiler successfully operated on the Linux kernel, demonstrating Opus 4.6's capabilities in complex software engineering. This experiment highlights significant advancements in autonomous software development, suggesting that AI agents can now independently execute sophisticated programming tasks. As reported by Anthropic, these findings have major implications for the future of software engineering, potentially reducing development time and costs while expanding business opportunities for automated coding and systems integration.

According to @godofprompt on Twitter, Claude has launched 'agent teams' on Claude Code, allowing users to deploy multiple AI agents that coordinate autonomously and work in parallel. This feature, currently in research preview, enables the efficient handling of complex tasks that can be divided and managed independently, driving new business opportunities in workflow automation and scalable AI solutions. As reported by Claude, this development marks a significant step in advancing autonomous agent collaboration for enterprise productivity.

According to Anthropic (@AnthropicAI), Opus 4.6 was assigned the task of building a C compiler using coordinated agent teams, with minimal human intervention. After two weeks, the resulting compiler was able to successfully run on the Linux kernel. This experiment demonstrates the significant potential of autonomous agent-driven software development and points to a future where advanced AI systems like Opus 4.6 can independently create complex, production-level tools. The blog highlights key lessons about coordination, efficiency, and the transformative business opportunities in deploying AI agents for large-scale software engineering tasks.

According to Lydia Hallie on Twitter, Claude Code has introduced a new experimental feature called Teams, also referred to as Agent Swarms, which allows a lead agent to delegate tasks to multiple AI teammates working in parallel. This update enables agents to research, debug, and build collaboratively, greatly enhancing productivity and workflow automation. As reported by Lydia Hallie, the Teams feature is currently in research preview and may require significant token usage. Users can enable Teams in their settings.json and access further documentation on the official Claude Code site. This development signals growing business opportunities for companies looking to streamline operations with collaborative AI agents.

According to @claudeai, Claude Code has introduced agent teams, allowing users to deploy multiple autonomous agents that coordinate and work in parallel. This feature, currently in research preview, is designed for tasks that can be divided and handled independently, streamlining workflows and improving efficiency for complex projects, as reported on Twitter.