The command line is being reborn, transformed by LLMs into an accessible interface for everyone. This presentation chronicles the journey of creating Gemini CLI at Google that went from concept to over a million users in months.

We’ll explore how Gemini CLI wrote most of its own code and shifted the team from developers to “conductors” of AI. Finally we’ll dig through how we’ve navigated the challenges of hyper-growth and a massive open-source community. This session offers a transparent look at the new bottlenecks that emerge when development is instantaneous, where communication, not coding, is the primary challenge. Attendees will leave with key insights on the shift from a “10x” to “100x” engineering mindset and the critical importance of UX in building the next generation of AI-native products.