The Claude Code creator’s workflow is a real-time strategy game for coding

When the creator of the world’s most advanced coding agent shares his personal workflow, you don’t scroll past. Boris Cherny, the guy behind <a href="https://code.allwinchina.org/ai-tools/claude-code/" title="Claude Code review”>Claude Code at Anthropic, dropped a thread on X last week that the engineering community is still picking apart. What started as a casual peek into his terminal setup has turned into something of a manifesto on how software should be built in 2026.

Jeff Tang, a well-known developer voice, put it bluntly: “If you’re not reading the Claude Code best practices straight from its creator, you’re behind as a programmer.” Another observer, Kyle McNease, went further, saying Anthropic is “on fire” and potentially facing “their ChatGPT moment.”

The wild part? Cherny’s workflow is deceptively simple. But it lets a single human operate with the output of a small engineering department. One user on X described the experience after implementing Cherny’s setup as “feels more like Starcraft” than traditional coding. That’s the shift we’re talking about — from typing syntax to commanding autonomous units.

Running five AI agents at once

The most striking thing about Cherny’s disclosure is that he doesn’t code linearly. The traditional “inner loop” of development — write a function, test it, move to the next — is gone. Instead, Cherny acts like a fleet commander.

“I run 5 Claudes in parallel in my terminal,” he wrote. “I number my tabs 1-5, and use system notifications to know when a Claude needs input.”

He uses iTerm2 system notifications to manage five simultaneous work streams. While one agent runs a test suite, another refactors a legacy module, and a third drafts documentation. He also has “5-10 Claudes on claude.ai” running in his browser, with a “teleport” command to hand off sessions between the web and his local machine.

This validates the “do more with less” strategy Anthropic President Daniela Amodei talked about earlier this week. While competitors like OpenAI are building trillion-dollar infrastructure, Anthropic is proving that better orchestration of existing models can yield exponential productivity gains. I find that refreshing — it’s not always about bigger hardware.

Why he uses the slowest, smartest model

Here’s a counterintuitive move from someone who clearly values speed: Cherny exclusively uses Anthropic’s heaviest, slowest model, Opus 4.5.

“I use Opus 4.5 with thinking for everything,” he explained. “It’s the best coding model I’ve ever used, and even though it’s bigger & slower than Sonnet, since you have to steer it less and it’s better at tool use, it is almost always faster than using a smaller model in the end.”

This is a critical insight for anyone managing engineering teams. The bottleneck in modern AI development isn’t token generation speed — it’s the human time spent correcting the AI’s mistakes. Cherny’s approach suggests paying the “compute tax” for a smarter model upfront eliminates the “correction tax” later. I’ve seen this play out in practice: faster models often produce more bugs, and debugging takes longer than waiting for a better answer.

One file that turns every mistake into a lesson

Cherny also addressed the problem of AI amnesia. Standard LLMs don’t remember your company’s coding style or architectural decisions from one session to the next. His solution is elegant: a single file named CLAUDE.md checked into the git repository.

“Anytime we see Claude do something incorrectly we add it to the CLAUDE.md, so Claude knows not to do it next time,” he wrote.

This turns the codebase into a self-correcting organism. When a human 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. This is the kind of practical innovation that actually moves the needle.

Slash commands automate the tedious stuff

The “vanilla” workflow that observers 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 daily. Instead of manually typing git commands, writing a commit message, and opening a pull request, the agent handles the entire flow. It’s the kind of automation that feels obvious once you see it, but most teams never get around to building.

This is where the real productivity gains come from. Not from bigger models or faster hardware, but from smarter orchestration and a willingness to invest in workflow infrastructure. Cherny’s approach is a masterclass in that philosophy.