Private AI

Every prompt you send to a cloud AI lives on someone else's server. Sometimes that's fine. Sometimes it isn't.

Private AI is a step-by-step guide to running Claude Code on your own machine, with no API bill and no data leaving your laptop. By the end of Chapter 6 you'll have Claude Code wired to a free local model. By the end of the book, you'll know which models suit which jobs, when to stay local and when the cloud is the right call, and how to fix the four problems that actually break local-AI setups.

What's inside:

• Hardware reality: what you can run on the laptop you already own - and when honesty says you can't
• A fifteen-minute Ollama install for Mac, Windows, and Linux
• Picking your first local model without drowning in benchmark blogs
• Two ways to connect Claude Code to Ollama, with the trade-offs of each
• Privacy workflows for legal, medical, financial, and consulting work
• Hybrid setups that keep sensitive drafts local and reach for the cloud only when needed
• A real project: a private research assistant for source-protected work
• Troubleshooting the four problems that break most local-AI setups

Written for non-developers, freelancers, and professionals in regulated industries who already use AI but cannot - or no longer want to - send their work to someone else's servers. No prior local-AI experience needed. Every command was tested on the stated operating system before publication.

Book 2 of The Practical AI Series. Each book stands alone.