skinny.

Latest / AI Fire Daily

#34 Robin: Stop the API Bleeding - Running Claude Code Locally with Gemma 4 and LM Studio

Got something to say to this podcast? Send a voicemail, and you might hear yourself on the next episode!

Every time you hit "Enter" on a coding agent, you’re basically swiping your credit card. But in 2026, the real pros aren't just spending tokens—they’re optimizing them. Today, we’re breaking down the "Zero-Token Developer" stack: how to run Claude Code entirely on your local machine using Gemma 4 and LM Studio.

We explore the reality of "Hand-off Engineering"—the strategy of using top-tier models like Claude 3.7 for the high-level architecture, then handing the repetitive "muscle work" to…

The skinny

The skinny isn't ready yet — notes appear once the transcript is processed.