Anthropic’s Code with Claude showed off coding’s future—whether you like it or not
Anthropic’s two-day event, Code with Claude, held in London on May 19, showcased how AI tools like Claude Code are reshaping software development by automating code generation and review. Engineers at the event revealed that much of Anthropic’s own software is now written by Claude, with developers increasingly relying on AI to handle coding tasks without manual oversight.
Anthropic hosted its second annual Code with Claude event in London on May 19, highlighting the growing role of AI in software development. The event, attended by developers actively using tools like Claude Code, demonstrated how AI-generated code is now commonplace in professional workflows. Jeremy Hadfield, an Anthropic engineer, noted that nearly half the attendees had shipped pull requests entirely written by Claude, with many admitting they didn’t review the code at all. Claude Code, powered by Anthropic’s latest AI models like Claude 4.6 and 4.7, now automates not just code generation but also error correction and testing. Boris Cherny, head of Claude Code, emphasized a shift where developers prompt AI to refine its own work, reducing human intervention. Ravi Trivedi described this as ‘letting it cook,’ with AI handling iterative debugging and adjustments behind the scenes. A new feature called ‘dreaming’ allows Claude Code agents to document task-related insights, enabling future agents to learn from past errors and optimize workflows. This self-improving system consolidates patterns across tasks, improving efficiency in codebases over time. Anthropic’s event included presentations from companies leveraging Claude Code, reinforcing its adoption as a standard tool. The company’s goal is to push automation further, with human developers increasingly acting as overseers rather than hands-on coders. Similar trends are observed at competitors like OpenAI, Google, and Microsoft, where AI-driven coding tools are becoming integral to development processes.
This content was automatically generated and/or translated by AI. It may contain inaccuracies. Please refer to the original sources for verification.