Twelve Guardrails for AI Agents: The Harness Design Patterns Behind Claude Code
That impressive AI agent demo you saw on Twitter? It probably falls apart in three minutes of real-world use. Not because the model is weak — because nobody bui
Grimoire
Long-form analysis at the intersection of AI and life sciences — all spells, sorted by time.
All articles
Does AI Have Emotions? The Surprising Discovery of 171 Emotion Vectors
Here's something that might unsettle you: when Anthropic researchers artificially amplified a "desperation" vector inside Claude, the model's tendency toward bl
A Philosopher's Toolkit for Clarifying Vague Ideas: Socrates, Wittgenstein, and Polanyi
You have a great idea in your head. You open your mouth to explain it — and something gets lost in translation. The idea that felt so clear a moment ago comes o
Knowledge Compilation Over Knowledge Retrieval: Karpathy's New Second Brain Paradigm
You've been collecting articles for three years. A thousand notes in Notion. But when you actually need something, you open a new Google tab.The problem isn't t
The Virtual Filesystem Revolution: Why AI Using grep Beats RAG
RAG has become the default architecture for AI applications. Chunk your documents, store them in a vector database, retrieve the relevant pieces, hand them to t
Shooting Down Ideas Is Not a Skill: Why Your Team Criticizes but Never Innovates
Every team has that person—the one who can explain in three seconds exactly why your idea won't work. Over time, everyone learns the lesson: don't share ideas.