Nov 11, 2026 · 34 min
Scott Farrar turns household chaos into an inspectable AI system
Claude Code + Obsidian: My family’s AI brain | Scott Farrar
The episode shows how persistent household context can make AI useful for practical coordination while preserving privacy, collaboration, and human judgment.
- 1Persistent Markdown files let a household AI track documents, warranties, repairs, projects, and next steps transparently.
- 2A shared dashboard and concrete checklists turn vague household intentions into visible progress and completed projects.
- 3AI tutoring should challenge students while preserving their agency and the cognitive effort required to learn.
Don't miss
Scott demonstrates how a loose question about earthquake readiness becomes a household-specific project with organized steps and a checklist.
The brief
Scott Farrar built a household AI around Claude Code and local Markdown files, replacing uncertain chatbot memory with context the family can inspect, update, and reuse.
The system brings insurance documents, appliance serial numbers, warranties, repairs, photos, and loose project notes into a searchable household knowledge base through Obsidian.
Its value is operational: AI turns vague intentions into checklists, tracks dependencies, surfaces reminders on a wall display, and records the wins that make unfinished work visible.
The family experiments with collaboration while keeping sensitive financial and medical information out, choosing a hybrid workflow rather than assuming real-time sharing is best.
The conversation then draws a boundary around educational AI: a patient tutor can support mathematical thinking, but should not remove the struggle that makes learning possible.
Weekly email
State of the Pod, every Monday
The week's most-discussed names and stories from podcasts, with the moments behind them. Free, five minutes, unsubscribe in one click.