In the field of artificial intelligence, "prompt engineering" is rapidly giving way to "loop engineering." According to Business Insider, Boris Cherny, the creator of Claude Code, recently stated that he no longer writes prompts—instead, he lets AI agents complete tasks autonomously through loop mechanisms. "It's an agent that prompts Claude," Cherny explained in the interview. "I don't even write prompts anymore." He considers loops and similar mechanisms among the achievements he is most proud of after a decade of work.
This trend is not isolated. OpenAI engineer Peter Steinberger, creator of the viral open-source project OpenClaw, publicly reminded users on X: "Monthly reminder: You should no longer manually write prompts for coding agents." Claire Vo, founder of ChatPRD, summarized: "The core message is to remind people that you don't need to type prompts with human fingers—the agent can do the work for you." The essence of loop engineering is to create automated feedback loops that allow AI agents to self-drive based on context, significantly reducing human intervention.
