Carousel Studio

Cortex

@0xCortexl

Anthropic pays their engineers $80,000 a month. Those engineers now merge 8x more code per day than they did a year ago. Not because the model got smarter. Because they stopped prompting Claude and started building systems that prompt it for them.

Apply Image

Cortex

@0xCortexl

Brian Cherny, head of Claude Code at Anthropic: "I don't prompt Claude anymore. I have loops running that prompt Claude and figure out what to do. My job is to write loops."

Cortex

@0xCortexl

Most developers read that and have no idea what it means in practice. By the end of this article you will - and you'll have everything you need to build your first one this weekend.

Cortex

@0xCortexl

<b>What a loop actually is</b>

Cortex

@0xCortexl

A prompt is a single instruction. A loop is a goal the AI keeps working toward until it gets there.

Cortex

@0xCortexl

The difference is who does the driving. With a prompt you push Claude through every step manually. With a loop you define the goal once and the system runs the full cycle on its own - finding the work, executing it, checking the result and deciding what comes next. All without you in the chair.

Cortex

@0xCortexl

<pre><code lang="">Prompt | you type, Claude answers, you type again Loop | system finds work, Claude executes, | system checks result, repeats until done</code></pre>

Cortex

@0xCortexl

Every real loop has five stages:

Cortex

@0xCortexl

<pre><code lang="">Discover | find what needs doing Plan | decide how to do it Execute | do the work Verify | check against the goal Iterate | not done? feed result back in and go again</code></pre>

Cortex

@0xCortexl

Three of these do all the real work.

Cortex

@0xCortexl

<b>Verify is the heart.</b> Without a real check on the result you don't have a loop - you have the agent agreeing with itself on repeat. The check has to be objective. A test that passes or fails. A build that compiles or doesn't. A linter that returns zero or non-zero. Not a second agent asked to "review." Two optimists agreeing is not verification.

Cortex

@0xCortexl

<b>State is what makes the loop learn.</b> Each pass the AI has to remember what it already tried or it repeats the same mistake forever. A real loop keeps a record outside the conversation - what is done, what failed, what is next. The agent forgets between sessions. The file doesn't.

Cortex

@0xCortexl

<b>A stop condition is what keeps it sane.</b> A loop with no exit runs until it succeeds, breaks or drains your budget. Every serious loop has two ways to stop - success, and a hard limit. Without this you've built a machine that bills you in silence.

Cortex

@0xCortexl

<b>The 4-condition test before you build anything</b>

Cortex

@0xCortexl

A loop earns its cost only when all four of these are true. Miss one and keep it as a manual prompt.

Cortex

@0xCortexl

<pre><code lang="">Condition 1 | the task repeats at least weekly | less than weekly - setup cost never pays back Condition 2 | something can automatically reject bad output | a test, build, linter, type check | no automated gate - you're back reviewing every diff Condition 3 | the agent can do the work end to end | not hand half of it back to you mid-task Condition 4 | done is objective, not a judgment call | if quality is a matter of taste, a human still wins</code></pre>

Cortex

@0xCortexl

Good first loops:

Cortex

@0xCortexl

<pre><code lang="">CI failure triage | nightly, classify failures, draft fixes Dependency bumps | weekly, scan updates, open PRs Lint-and-fix passes | on every PR, apply style fixes automatically Issue-to-PR drafts | on codebases with strong test coverage</code></pre>

Cortex

@0xCortexl

Bad first loops - keep a human in the chair:

Cortex

@0xCortexl

<pre><code lang="">Architecture rewrites | judgment call, loop will drift Auth or payments code | irreversible mistakes too expensive Production deploys | needs human approval gate Vague product work | done is not objective</code></pre>