S-TIER guide
Claude Code Team Tips — Boris Cherny's 10 habits for parallel, plan-first agent work
100k ft — What is this?
Claude Code Team Tips — Boris Cherny's 10 habits for parallel, plan-first agent work
50k ft — Why care?
Boris Cherny's thread of tips from the Claude Code team: 3-5 git worktrees with parallel sessions, plan mode for every complex task, a CLAUDE.md you update after each correction, skills committed to git, Slack MCP bug fixing, subagents, a BigQuery skill for analytics and learning output styles. It ends with a ready-to-paste CLAUDE.md workflow template (plan, subagents, lessons file, verify before done).
10k ft — Key takeaways
• Adopt 3-5 parallel worktrees per feature as default Claude Code workflow
• End every correction with 'update your CLAUDE.md so you don't make that mistake again'.
• Run 3-5 worktrees each with its own Claude session; name them and alias them for fast switching.
• Start complex work in plan mode and re-plan when things drift instead of pushing on.
• This matches Victor's dispatching-parallel-agents skill
5k ft — Tools & techniques
BigQuery, Claude, Claude Code, Ghostty, Slack, git, tmux
CLAUDE.md rules, MCP server, git worktrees, parallel git worktrees, permission hooks, plan mode, slash commands, subagent orchestration, subagents, workflow automation
1k ft — Deep-dive / try it
💻🧠 The team behind Claude Code just shared they actually uses it internally and why many new users struggle at the start.
In a recent thread on X, Boris Cherny explained that the team’s workflow looks very different from how most people approach the tool.
Instead of running one session, they often spin up 3 to 5 git worktrees in parallel, each with its own Claude session, which they describe as a major productivity unlock.
They also begin every complex task in plan mode and focus heavily on writing a strong plan before implementation. If something breaks or drifts off course, they go back to planning rather than forcing a fix.
Another key habit is updating Claude’s own rules after mistakes so the same errors do not repeat. Over time, this lowers the mistake rate and improves reliabilit…
Appears in
- Guide: The Self-Improving AI Workspace: Claude Code That Gets Smarter Every Session
- Deeper dive: The Self-Improving AI Workspace: Claude Code That Gets Smarter Every Session
- Guide: The Claude Code Power User's Toolkit: From Token Waste to Terminal Dominance
- Deeper dive: The Claude Code Power User's Toolkit: From Token Waste to Terminal Dominance
- Guide: Claude Mastery: From First Install to Advanced Agent Architecture
- Deeper dive: Claude Mastery: From First Install to Advanced Agent Architecture
- Guide: Mission Control for 80 Agents — Conway's law meets game UI
- Deeper dive: Mission Control for 80 Agents — Conway's law meets game UI
- Guide: Claude Code Parallel Worktrees