Strategies for Effective Human-AI Teamwork
Based on insights from Boris Cherny (@bcherny), Creator of Claude Code, and the Claude Code team
This guide outlines proven strategies for collaborating effectively with Claude Code as a team member. These tips come directly from how the Claude Code team works together with Claude in production environments.
Remember: there is no one right way to use Claude Code -- everyone's setup is different. You should experiment to see what works for you!
The key is treating Claude Code as a collaborative partner rather than just a tool -- find the workflow that enables the best teamwork between you and Claude.
The single biggest productivity unlock according to the team.
- Spin up 3-5 git worktrees at once, each running its own Claude session in parallel
- Most of the Claude Code team prefers worktrees over multiple git checkouts
- This is why native support for worktrees was built into the Claude Desktop app!
- Name your worktrees and set up shell aliases (za, zb, zc) so you can hop between them in one keystroke
- Have a dedicated "analysis" worktree that's only for reading logs and running BigQuery
Learn more: Parallel Claude Code Sessions with Git Worktrees
Pour your energy into the plan so Claude can 1-shot the implementation.
Multi-Claude Review:
- Have one Claude write the plan
- Spin up a second Claude to review it as a staff engineer
When Things Go Wrong:
- The moment something goes sideways, switch back to plan mode and re-plan
- Don't keep pushing forward with a broken approach
- Explicitly tell Claude to enter plan mode for verification steps, not just for the build
Claude is eerily good at writing rules for itself.
After every correction, end with:
"Update your CLAUDE.md so you don't make that mistake again."
- Ruthlessly edit your CLAUDE.md over time
- Keep iterating until Claude's mistake rate measurably drops
- One engineer tells Claude to maintain a notes directory for every task/project, updated after every PR, then points CLAUDE.md at it
Reuse skills across every project.
- If you do something more than once a day, turn it into a skill or command
- Build a
/techdebtslash command and run it at the end of every session to find and kill duplicated code - Set up a slash command that syncs 7 days of Slack, GDrive, Asana, and GitHub into one context dump
- Build analytics-engineer-style agents that write dbt models, review code, and test changes in dev
Here's how the team does it:
Slack Integration:
- Enable the Slack MCP
- Paste a Slack bug thread into Claude and just say "fix."
- Zero context switching required
CI/CD:
- Just say "Go fix the failing CI tests."
- Don't micromanage how
Distributed Systems:
- Point Claude at docker logs to troubleshoot distributed systems
- It's surprisingly capable at this
Make Claude be your reviewer:
- Say: "Grill me on these changes and don't make a PR until I pass your test."
- Say: "Prove to me this works" and have Claude diff behavior between main and your feature branch
After a mediocre fix:
"Knowing everything you know now, scrap this and implement the elegant solution"
- Write detailed specs and reduce ambiguity before handing work off
- The more specific you are, the better the output
Terminal:
- The team loves Ghostty!
- Multiple people praise its synchronized rendering, 24-bit color, and proper unicode support
Status Line:
- Use
/statuslineto customize your status bar to always show:- Context usage
- Current git branch
Tab Management:
- Color-code and name your terminal tabs
- Many use tmux — one tab per task/worktree
Voice Dictation:
- Use voice dictation (hit fn x2 on macOS)
- You speak 3x faster than you type
- Your prompts get way more detailed as a result
a. Request More Compute:
- Append "use subagents" to any request where you want Claude to throw more compute at the problem
b. Keep Context Clean:
- Offload individual tasks to subagents to keep your main agent's context window clean and focused
c. Permission Automation:
- Route permission requests to Opus 4.5 via a hook
- Let it scan for attacks and auto-approve the safe ones
- Learn more about permission hooks
Ask Claude Code to use the "bq" CLI to pull and analyze metrics on the fly.
- The team has a BigQuery skill checked into the codebase
- Everyone uses it for analytics queries directly in Claude Code
- Boris personally hasn't written a line of SQL in 6+ months
This works for any database that has a CLI, MCP, or API.
A few tips from the team to use Claude Code for learning:
- Enable the "Explanatory" or "Learning" output style in
/config - Have Claude explain the why behind its changes
- Have Claude generate a visual HTML presentation explaining unfamiliar code
- It makes surprisingly good slides!
- Ask Claude to draw ASCII diagrams of new protocols and codebases to help you understand them
Build a spaced-repetition learning skill:
- You explain your understanding
- Claude asks follow-ups to fill gaps
- Claude stores the result for future reference
- Start small: Pick 2-3 tips that resonate with your workflow
- Experiment: Try different approaches to see what works for you
- Iterate: Continuously refine your setup based on what improves your productivity
- Share: Contribute your own discoveries back to the community
Last updated: February 2025 Source: Twitter thread by Boris Cherny (@bcherny)