One year ago, we made Claude Code generally available. What started as an internal project—an agentic coding tool that runs in your terminal—is now used by developers and organizations worldwide.
Boris Cherny (Head of Claude Code) and Cat Wu (Head of Product, Claude Code) look back on the Claude Code's first year, from a Slack demo that got two reactions to engineering teams deploying it across entire codebases. They cover best practices for verification, the thinking behind auto mode, their favorite routines and loops, Claude Code's adoption beyond engineering, the rise of context minimalism, and how to build for the AI exponential.
0:00 - The origins and evolution of Claude Code 1:10 - How to make Claude good at verification 3:14 - Roles merging: Claude Code beyond engineers 4:48 - Using routines for CI, code review, and more 6:43 - Boris' go-to feature: auto mode 8:10 - Securing auto mode: red teaming and evals 10:24 - Why loop is the next leap 11:06 - How engineering orgs and responsibilities are changing 13:30 - Is the future product or engineering? 14:20 - Working with hundreds of agents: using agent view, voice mode, and Remote Control 16:05 - From context engineering to context minimalism 17:17 - What's next for Claude Code
Learn more about Claude Code: https://code.claude.com/docs/en/overview
Follow ClaudeDevs on X for product updates and best practices from the Claude Code team: https://x.com/ClaudeDevs
