When the creator of the world's most advanced coding agent speaks, Silicon Valley doesn't just listen — it takes notes.
For the past week, the engineering community has been dissecting a thread on X from Boris Cherny, the creator and head of Claude Code at Anthropic. What began as an off-the-cuff sharing of his personal terminal setup has spiraled right into a viral manifesto on the long run of software development, with industry insiders calling it a watershed moment for the startup.
"In case you're not reading the Claude Code best practices straight from its creator, you're behind as a programmer," wrote Jeff Tang, a outstanding voice within the developer community. Kyle McNease, one other industry observer, went further, declaring that with Cherny's "game-changing updates," Anthropic is "on fire," potentially facing "their ChatGPT moment."
The joy stems from a paradox: Cherny's workflow is surprisingly easy, yet it allows a single human to operate with the output capability of a small engineering department. As one user noted on X after implementing Cherny's setup, the experience "feels more like Starcraft" than traditional coding — a shift from typing syntax to commanding autonomous units.
Here is an evaluation of the workflow that’s reshaping how software gets built, straight from the architect himself.
How running five AI agents without delay turns coding right into a real-time strategy game
Probably the most striking revelation from Cherny's disclosure is that he doesn’t code in a linear fashion. In the standard "inner loop" of development, a programmer writes a function, tests it, and moves to the following. Cherny, nonetheless, acts as a fleet commander.
"I run 5 Claudes in parallel in my terminal," Cherny wrote. "I number my tabs 1-5, and use system notifications to know when a Claude needs input."
By utilizing iTerm2 system notifications, Cherny effectively manages five simultaneous work streams. While one agent runs a test suite, one other refactors a legacy module, and a 3rd drafts documentation. He also runs "5-10 Claudes on claude.ai" in his browser, using a "teleport" command at hand off sessions between the net and his local machine.
This validates the "do more with less" strategy articulated by Anthropic President Daniela Amodei earlier this week. While competitors like OpenAI pursue trillion-dollar infrastructure build-outs, Anthropic is proving that superior orchestration of existing models can yield exponential productivity gains.
The counterintuitive case for selecting the slowest, smartest model
In a surprising move for an industry obsessive about latency, Cherny revealed that he exclusively uses Anthropic's heaviest, slowest model: Opus 4.5.
"I take advantage of Opus 4.5 with pondering for all the pieces," Cherny explained. "It's the most effective coding model I've ever used, and though it's larger & slower than Sonnet, since you’ve to steer it less and it's higher at tool use, it is sort of at all times faster than using a smaller model ultimately."
For enterprise technology leaders, it is a critical insight. The bottleneck in modern AI development isn't the generation speed of the token; it’s the human time spent correcting the AI's mistakes. Cherny's workflow suggests that paying the "compute tax" for a wiser model upfront eliminates the "correction tax" later.
One shared file turns every AI mistake right into a everlasting lesson
Cherny also detailed how his team solves the issue of AI amnesia. Standard large language models don’t "remember" an organization's specific coding style or architectural decisions from one session to the following.
To handle this, Cherny's team maintains a single file named CLAUDE.md of their git repository. "Anytime we see Claude do something incorrectly we add it to the CLAUDE.md, so Claude knows to not do it next time," he wrote.
This practice transforms the codebase right into a self-correcting organism. When a human developer reviews a pull request and spots an error, they don't just fix the code; they tag the AI to update its own instructions. "Every mistake becomes a rule," noted Aakash Gupta, a product leader analyzing the thread. The longer the team works together, the smarter the agent becomes.
Slash commands and subagents automate essentially the most tedious parts of development
The "vanilla" workflow one observer praised is powered by rigorous automation of repetitive tasks. Cherny uses slash commands — custom shortcuts checked into the project's repository — to handle complex operations with a single keystroke.
He highlighted a command called /commit-push-pr, which he invokes dozens of times each day. As a substitute of manually typing git commands, writing a commit message, and opening a pull request, the agent handles the bureaucracy of version control autonomously.
Cherny also deploys subagents — specialized AI personas — to handle specific phases of the event lifecycle. He uses a code-simplifier to wash up architecture after the essential work is completed and a verify-app agent to run end-to-end tests before anything ships.
Why verification loops are the actual unlock for AI-generated code
If there may be a single reason Claude Code has reportedly hit $1 billion in annual recurring revenue so quickly, it is probably going the verification loop. The AI is just not only a text generator; it’s a tester.
"Claude tests each change I land to claude.ai/code using the Claude Chrome extension," Cherny wrote. "It opens a browser, tests the UI, and iterates until the code works and the UX feels good."
He argues that giving the AI a technique to confirm its own work — whether through browser automation, running bash commands, or executing test suites — improves the standard of the by "2-3x." The agent doesn't just write code; it proves the code works.
What Cherny's workflow signals concerning the way forward for software engineering
The response to Cherny's thread suggests a pivotal shift in how developers take into consideration their craft. For years, "AI coding" meant an autocomplete function in a text editor — a faster technique to type. Cherny has demonstrated that it may possibly now function as an operating system for labor itself.
"Read this if you happen to're already an engineer… and need more power," Jeff Tang summarized on X.
The tools to multiply human output by an element of 5 are already here. They require only a willingness to stop pondering of AI as an assistant and begin treating it as a workforce. The programmers who make that mental leap first won't just be more productive. They'll be playing a wholly different game — and everybody else will still be typing.
