Claude Code’s creator shared his workflow, and it’s basically playing Starcraft with five AI agents

When the guy who built the most advanced coding agent in the world casually drops his personal workflow on X, you pay attention. Boris Cherny, creator and head of <a href="https://edu.allwinchina.org/ai-tools/claude-code/" title="Claude Code review”>Claude Code at Anthropic, did exactly that last week, and the engineering community has been picking it apart ever since.

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

What’s interesting is that Cherny’s setup is deceptively simple. But it lets a single human operate with the output capacity of a small engineering team. One user on X said implementing his setup “feels more like Starcraft” than traditional coding. You’re not typing syntax anymore — you’re commanding autonomous units.

Five agents, one terminal, no waiting

The most striking thing about Cherny’s disclosure is that he doesn’t code linearly. The traditional “inner loop” — write a function, test it, move on — is dead to him. He acts like 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.”

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 runs 5-10 Claudes on claude.ai in his browser, using a “teleport” command to hand off sessions between web and local machine.

This validates the “do more with less” strategy Anthropic President Daniela Amodei talked about recently. While OpenAI chases trillion-dollar infrastructure, Anthropic is showing that smarter orchestration of existing models can deliver exponential productivity gains.

The case for the slowest, smartest model

Here’s the counterintuitive part. In an industry obsessed with latency, 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 building with AI. The bottleneck isn’t token generation speed — it’s the human time spent correcting the AI’s mistakes. Paying the “compute tax” for a smarter model upfront eliminates the “correction tax” later. I’ve seen this play out in my own work: the faster model that requires constant steering is almost always slower in wall-clock time.

One file that turns every mistake into a permanent lesson

Standard LLMs don’t remember your coding style or architectural decisions from session to session. Cherny’s team solved this with a single file: CLAUDE.md, stored in their 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 gets.

Slash commands that automate the boring stuff

Cherny’s “vanilla” workflow is powered by rigorous automation of repetitive tasks. He uses slash commands — custom shortcuts checked into the project’s repository — to handle complex operations with a single keystroke.

His go-to is /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 everything. This isn’t revolutionary in concept, but the execution matters: these commands are version-controlled and shared across the team, so everyone benefits from the accumulated automation.

What this means for the rest of us

Cherny’s workflow isn’t just a curiosity — it’s a blueprint. The era of the solo developer writing every line of code is ending. The future is one human orchestrating multiple AI agents, each handling different parts of the development process.

The key takeaways are straightforward:

1. Parallelize your agents. Don’t wait for one task to finish before starting another.
2. Use the smartest model you can afford. The time you save on corrections is worth the extra latency.
3. Build a shared memory system. A single CLAUDE.md file that accumulates rules and preferences turns your codebase into a learning system.
4. Automate everything repetitive. Slash commands for common workflows eliminate friction.

Is this the “ChatGPT moment” for Anthropic? I’m not sure. But Cherny’s disclosure shows that the real competitive advantage isn’t just the model — it’s how you orchestrate it. And right now, Anthropic is showing the rest of the industry how it’s done.