Why search alone does not create memory—and how source layers, agent rules, and deterministic checks...
A Markdown/Git knowledge base separates immutable sources from revisable conclusions; agents update synthesis with new evidence. RAG/search retrieves fragments without memory; maintained wikis preserve audit trails and prevent rediscovering contradictions.
Why search alone does not create memory—and how source layers, agent rules, and deterministic checks keep a knowledge base useful over time
My AI knowledge base starts with a directory tree, not a search box:
10-inbox rough, unverified capture
20-sources evidence that cannot be silently rewritten
30-knowledge synthesis that can change with new evidence
When you're a solo founder and lead architect, your knowledge base is your most valuable asset. Lose...
How we measured a markdown knowledge graph as agent memory — lost to grep, rebuilt our search and...
Claude is incredibly good at reasoning. But reasoning is only as useful as the context available to...
I Built a Local Knowledge Graph That Tells AI Agents "What Breaks If I Change This?" Every...
Why I built a knowledge arbitration engine instead of just another memory layer If you've...
Four serious AI-memory projects build elaborate machinery around an extraction step they each treat as solved. I built the fifth…
