# nunez-cherny-claude-code-workflow-venturebeat-2026-01-05

## Veille

Boris Cherny, creator of Claude Code - workflow with 5 parallel agents, Opus 4.5, CLAUDE.md

## Titre Article

The creator of Claude Code just revealed his workflow, and developers are losing their minds

## Date

2026-01-05

## URL

https://venturebeat.com/technology/the-creator-of-claude-code-just-revealed-his-workflow-and-developers-are

## Keywords

Claude Code, Boris Cherny, Anthropic, workflow, parallel agents, Opus 4.5, CLAUDE.md, subagents, slash commands, verification, developer productivity

## Authors

Michael Nuñez

## Ton

Profile: Tech journalism article, enthusiastic and analytical register, viral news perspective.
Style: Engaging tone with a dramatic opening ("Silicon Valley doesn't just listen — it takes notes"). Structured into clear thematic sections. Recurring gaming metaphor (Starcraft, "fleet commander", "real-time strategy game"). Multiple quotes from Twitter reactions to amplify impact. Accessible vocabulary with integrated technical explanations. Target audience: developers, tech leaders, AI early adopters.

## Pense-betes

- **Source**: X thread by Boris Cherny (creator and head of Claude Code at Anthropic)
- **5 parallel agents**: tabs numbered 1-5 in iTerm2 with system notifications
- **5-10 additional Claudes** on claude.ai with a "teleport" command for web→terminal transfer
- **Exclusive model**: Opus 4.5 with thinking - "slower but fewer corrections = faster in the end"
- **CLAUDE.md**: single file in the git repo to correct the AI's recurring errors
- **Slash commands**: /commit-push-pr used dozens of times a day
- **Specialized subagents**: code-simplifier (architecture), verify-app (e2e tests)
- **Verification loop**: Claude tests via Chrome extension, opens a browser, iterates until functional
- **Quality improvement**: 2-3x thanks to automatic verification
- **Metaphor**: moving from "typing code" to "commanding autonomous units" (like Starcraft)
- **Key quote**: "AI as a workforce, not an assistant"
- **Context**: Claude Code reportedly reached $1B ARR

## RésuméDe400mots

Michael Nuñez reports on VentureBeat on the viral impact of an X thread by Boris Cherny, creator and head of Claude Code at Anthropic, detailing his development workflow. The developer community describes these revelations as a pivotal moment for Anthropic.

Cherny's workflow relies on running five Claude agents in parallel in his terminal. He numbers his iTerm2 tabs from 1 to 5 and uses system notifications to know when an agent requires intervention. While one agent runs a test suite, another refactors a legacy module, and a third writes documentation. He also launches 5 to 10 additional Claude sessions on claude.ai, using a "teleport" command to transfer sessions between the web and his local machine.

Unlike the industry's obsession with latency, Cherny exclusively uses Opus 4.5, Anthropic's heaviest and slowest model. His justification: although slower, this model requires fewer corrections and excels at tool use, ultimately making it faster than a lightweight model. The bottleneck is not generation speed but the human time spent correcting the AI's errors.

To solve the problem of LLM amnesia between sessions, the team maintains a single CLAUDE.md file in their git repository. Every time Claude makes a mistake, it is documented in this file to prevent its repetition. This practice transforms the codebase into a self-correcting organism where every error becomes a permanent rule.

Cherny uses slash commands - custom shortcuts versioned in the repository - to automate repetitive tasks. His /commit-push-pr command, used dozens of times a day, automatically handles versioning. He also deploys specialized subagents: a code-simplifier to clean up architecture and a verify-app for end-to-end testing.

The verification loop is the key innovation. Claude tests each change via the Chrome extension, opens a browser, tests the interface, and iterates until the code works and the UX is satisfactory. This automatic verification improves final quality by 2-3x.

The article concludes with a paradigm shift: AI is no longer an autocomplete assistant but an "operating system for work itself." Developers who adopt this vision will not just be more productive - they will be playing an entirely different game.

## GrapheDeConnaissance

- Boris Cherny —a_créé→ Claude Code (TECHNOLOGIE, 0.98)
- Boris Cherny —travaille_chez→ Anthropic (ORGANISATION, 0.98)
- Boris Cherny —utilise→ 5 agents parallèles (METHODOLOGIE, 0.97)
- Boris Cherny —utilise→ Opus 4.5 (TECHNOLOGIE, 0.95)
- CLAUDE.md —résout→ erreurs récurrentes de l'IA (CONCEPT, 0.93)
- Claude Code —mesure→ 1 milliard USD ARR (MESURE, 0.8)
- slash commands —permet→ automatisation des tâches répétitives développeur (CONCEPT, 0.9)
- boucle vérification —améliore→ qualité finale du code (2-3x) (CONCEPT, 0.88)
- Boris Cherny —utilise→ subagents spécialisés (code-simplifier, verify-app) (TECHNOLOGIE, 0.9)
- Opus 4.5 —réduit→ corrections nécessaires (CONCEPT, 0.88)

---
Canonical: https://www.thekb.eu/en/fiches/nunez-cherny-claude-code-workflow-venturebeat-2026-01-05/