Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
When the creator of the world’s most advanced coding agent speaks, Silicon Valley doesn’t just listen: it takes notes.
Over the past week, the engineering community has dissected a thread on Since Boris Chernythe creator and responsible for Claude Code has Anthropic. What started as an informal sharing of one’s personal terminal setup has turned into a viral manifesto on the future of software development, with industry insiders calling it a watershed moment for the startup.
"If you don’t read Claude Code’s best practices directly from its creator, you are behind as a programmer," wrote Jeff Tangan important voice in the developer community. Kyle McNeaseanother industry observer, went further, stating that with Cherny "revolutionary updates," Anthropic is "on fire," potentially confronted "their ChatGPT moment."
The excitement comes from a paradox: Cherny’s workflow is surprisingly simple, but it allows a single human to operate with the production capacity of a small engineering department. As one user on X noted after implementing Cherny’s setup, the experience "it looks more like Starcraft" than traditional coding – a shift from typing syntax to commanding autonomous units.
Here’s a workflow analysis that reshapes how software is built, straight from the architect himself.
How running five AI agents simultaneously turns coding into a real-time strategy game
The most striking revelation from Cherny’s revelation is that he does not code in a linear fashion. In the traditional "inner loop" During development, a programmer writes a function, tests it, and moves on to the next one. Cherny, however, acts as 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."
Using iTerm2 system notifications, Cherny efficiently manages five simultaneous workflows. While one agent runs a test suite, another refactors an existing module, and a third writes documentation. He also runs "5-10 Claudes on claude.ai" in your browser, using a "teleport" command to transfer sessions between the web and its local machine.
This validates the "do more with less" strategy articulated by Anthropic President Daniela Amodei earlier this week. As competitors like OpenAI continue to build billions of dollars worth of infrastructure, Anthropic is proving that superior orchestration of existing models can drive exponential productivity gains.
The Counterintuitive Case for Choosing the Slowest, Smartest Model
In a surprising move for an industry obsessed with latency, Cherny revealed that it uses Anthropic’s heaviest and slowest model exclusively: Opus 4.5.
"I use Opus 4.5 thinking of everything," CHRERY explain. "It’s the best coding model I’ve ever used, and even though it’s larger and slower than Sonnet, because it requires less direction and is better at using tools, it’s almost always faster than using a smaller model in the end."
For business technology leaders, this is essential information. The bottleneck in modern AI development is not the speed of token generation; that’s human time spent correcting AI errors. Cherny’s workflow suggests that paying the "calculate tax" for a smarter model eliminates from the start the "correction fee" later.
A shared file turns every AI mistake into a lifelong lesson
Cherny also explained how his team solves the problem of AI amnesia. Large standard language models don’t do this "remember" a company’s specific coding style or architectural decisions from session to session.
To solve this problem, the Cherny team maintains a single file named CLAUDE.md in their git repository. "Every time we see Claude doing something wrong, we add it to CLAUDE.md, so Claude knows not to do it next time." he wrote.
This practice transforms the code base into a self-correcting organism. When a human developer reviews a pull request and finds an error, they don’t just fix the code; they mark the AI to update its own instructions. "Every mistake becomes a rule," note Akash Guptaa product manager analyzing the feed. The longer the team works together, the smarter the agent becomes.
Slash commands and subagents automate the most tedious parts of development
THE "vanilla" The observer-lauded workflow is powered by rigorous automation of repetitive tasks. Cherny uses slash commands (custom shortcuts checked into the project repository) to handle complex operations with a single keystroke.
It highlighted a command called /commit-push-prwhich he invokes dozens of times a day. Instead of manually typing git commands, writing a commit message, and opening a pull request, the agent handles the version control bureaucracy autonomously.
Cherny also deploys sub-agents – specialized AI characters – to manage specific phases of the development lifecycle. It uses a code simplifyer to clean up the architecture after the main work is completed and an application verification agent to run end-to-end tests before anything is shipped.
Why verification loops are the real unlocker of AI-generated code
If there’s just one reason why Claude Code would have hit $1 billion in annual recurring revenue so fast, it’s probably the verification loop. AI is not just a text generator; it’s a tester.
"Claude tests every change I make to claude.ai/code using the Claude Chrome extension," Cherny wrote. "He opens a browser, tests the UI, and iterates until the code works and the UX looks nice."
He argues that giving AI a way to verify its own work – whether through browser automation, running bash commands, or running test suites – improves the quality of the end result by "2-3x." The agent doesn’t just write code; this proves that the code works.
What Cherny’s workflow signals about the future of software engineering
The reaction to Cherny’s thread suggests a crucial shift in the way developers think about their craft. For years, "AI coding" meant an autocomplete feature in a text editor – a faster way to type. Cherny demonstrated that it can now function as an operating system for the work itself.
"Read this if you’re already an engineer…and want more power," Jeff Tang summary on X.
The tools to increase human production fivefold are already there. They just need to be willing to stop thinking of AI as an assistant and start treating it like a workforce. Programmers who take this mental leap first will not only be more productive. They’ll play a completely different game – and everyone will keep typing.
