• Loops and loop engineering are all the talk in AI.
• OpenClaw creator Peter Steinberger said users "shouldn't be prompting coding agents anymore."
• Here's how automated tasks are changing the way AI agents are deployed.

For the most powerful voices in AI, it's all about being in the loop.

Claude Code creator Boris Cherny recently said he doesn't write his own AI prompts much anymore. Thanks to loops, he doesn't have to.

"It's an agent that prompts Claude," Cherny recently told CNBC, adding, "I don't write the prompt anymore. Claude writes the prompt, and now I'm talking to that new Claude that is kind of coordinating."

In the same interview, Cherny said that loops and a similar feature were examples of the kind of work he would be proudest of in a decade.

Cherny isn't the only one embracing "loop engineering." OpenAI engineer Peter Steinberger, the creator of the viral OpenClaw project, wrote a public reminder to users who are still writing out prompts for AI agents.

"Here's your monthly reminder that you shouldn't be prompting coding agents anymore," Steinberger wrote recently on X. "You should be designing loops that prompt your agents."

What are loops?

Loops are recurring systems that guide AI agents, so the user doesn't have to constantly write prompts themselves. An example is /goal, which instructs an AI tool like Anthropic's Claude Code or OpenAI's Codex to keep working until a task is completed, rather than requiring a prompt at each step.

Claire Vo, founder of ChatPRD and host of the "How I AI," might have summed it up best by saying, "it's really just reminding people that you don't have to use your human fingers to type in a prompt in order for your agent to do work on your behalf."

The days of directly prompting generative AI coding tools are "kind of over, or at least some think it's going to be," Addy Osmani, director of Google Cloud, wrote in his post explaining the concept.

Osmani wrote that loops need five components: automations, worktrees, skills, plugins and connectors, and sub-agents. The foundation is automation, because that's what ensures a loop can be repeated rather than a one-off event.

One of the most frequently suggested setups, especially when it comes to coding, is to split up your loop so that one agent writes the code and the other checks the final product.

"The model that wrote the code is way too nice grading its own homework," Osmani wrote.

Steinberger shared an example of a loop he uses: "Tell codex to maintain your repos, wake up every 5 minutes and direct work to threads. That makes it easy to parallelize+steer work as needed."

Here's a simple loop: Tell codex to maintain your repos, wake up every 5 minutes and direct work to threads. That makes it easy to parallelize+steer work as needed.

I use a orchestrator skill combined with my triage+autoreview+computer use skills, so some work can land… pic.twitter.com/0ASlWqkysW

— Peter Steinberger 🦞 (@steipete) June 11, 2026

When to use a loop

For the moment, the loop discourse is mostly focused on agentic coding. That doesn't mean loops are just for software engineers.

"This is the time for the manager," Vo said. "You are designing a job. And so just imagine that you're onboarding an employee. That employee could be an executive assistant, that employee could be a customer service agent, that employee could be a software engineer."

You might already be using a loop without knowing it.

"If you have used a scheduled task in ClaudeCowork, guess what, babe? You have written a loop," Vo said.

… And when not to

The biggest concern cited about loops is, by far, cost.

Running multiple agents with subagents on the latest frontier AI model is a great way to burn through your personal token budget or raise some eyebrows from your boss.

When users on X asked Steinberger about how he would modify his loop to be more budget-conscious, he responded: "Waking up and doing some API calls is fairly cheap, or opt for once per hour/day for lower token use."

At the same time, he also described himself as someone "the man with unlimited tokens." (Free tokens are one of the many perks of working at OpenAI.)

Osmani's advice is to spend only where necessary.

"Subagents do burn more tokens since each one does its own model and tool work, so spend them where a second opinion is worth paying for," he wrote.