Your Second Brain Is for Inference, Not Search

AI Mastermind | Knowledge Entrepreneurs Edition

“If you use Obsidian to look through all this, you’re going to get retrieval. But if you point your Claude on here, now you’re going to get inference on your knowledge.” — Lou

This Week in 30 Seconds

The Mac Disaster Lesson — Lou wiped and reinstalled from scratch after Apple blocked recovery from a moved home folder. Lesson for everyone: create a second admin account today, back up hidden config files (especially .claude/).
Spec-Driven Work Trees — Lou’s new development workflow: you write the spec, AI runs the full cycle (plan → code → test → PR). Work trees isolate experiments so anything that fails gets deleted, not merged.
Your Second Brain Is for Inference, Not Search — Kasimir’s Obsidian demo crystallized the key distinction: pointing Obsidian’s search at your vault gives you retrieval. Pointing Claude Code at it gives you inference. The links, tags, and taxonomy are the navigation map Claude follows to reason across your knowledge.
The Three-Layer Architecture — Optimal retrieval requires three layers: keyword search, knowledge graph traversal, and semantic similarity. Together, they approach 96.6% of RAG performance while adding temporal search that pure vector retrieval can’t do.
Scott’s One-Month Automation — Scott built an automated contact verification system in 2 days using a local Mistral 7B cron job. His EA went from a week of manual work to 10 exceptions to review. One automation a week = potentially saving a year of capacity by year-end.
Brand Before Publish — AI-generated HTML slides come in two defaults: purpley-blue or black-and-gold. Dirk’s workflow (Markdown → HTML) produces beautiful output. Lou’s extension: feed it your CSS and stand out in a homogeneous feed.

Three Layers of Memory

Kasimir runs three end-of-chat skills that fire automatically as context fills. One summarizes the conversation into Obsidian. One pushes the same summary into Notion. One sends it to Pinecone for retrieval. When he sits down the next day and Claude asks where to look, it queries across all three. Earlier this week he found himself rediscovering a Gemini gem he’d built a year ago and forgotten — exactly the failure mode the three-layer stack is designed to prevent.

Lou extended Kasimir’s pattern into a generalizable architecture. Tier one is the hot cache — a folder of files Claude greps directly, no semantic layer needed, optimized for what you’re actively working on this month. Tier two is the Karpathy-style wiki — atomic insight pages connected by wikilinks, ~800 to 900 entries in Lou’s current vault, structured for human reading and graph traversal. Tier three is the semantic memory — Pinecone-style vector retrieval over the long tail of everything you’ve ever written. Each tier serves a different question.

🔥 Hot Take: Your Second Brain Is for Inference, Not Search

Most people build a second brain to find things. The real payoff is using it to think with — feeding your accumulated knowledge into a model that can reason across all of it at once. Obsidian’s search tells you what you wrote. Claude Code on your vault tells you what it means in context of everything else you know. That’s a different tool entirely.

Next session: Thursday, May 14, 2026