How I Accidentally Built a Runaway AI
A half‑finished coding agent taught me three lessons: tokens are expensive, feedback loops matter, and guardrails aren’t optional
Nick Savage
I write mostly about agentic AI and LLMs as I work on Zettelgarden, a personal knowledge management system.
Live Last read · last published · next check
Latest posts
What I Learned Rebuilding Search Three Times
The evolution of Zettelgarden’s search: frontend hacks, brittle SQL, and finally a dedicated engine.
Facts, Arguments, Theses: Building AI Knowledge Retrieval on Meaning, Not Slices
Progress notes from Zettelgarden: experimenting with LLM pipelines that embed semantic units - theses, arguments, and facts - instead of raw text chunks
When AI Promises Speed but Delivers Debugging Hell
Lessons learned from building Codescribble with a little help (and a lot of frustration) from an AI
The Scenic Route to Software: Code Generation for Non-Traditional Developers
Observations from a self-taught developer building in public
Beyond RAG: Building a Knowledge Management System That Enhances Rather Than Replaces Thought
Preserving and augmenting human insight in an age of automated knowledge with Zettelgarden
From Vector Search to Entity Processing: Evolving Zettelgarden's Connection Engine
A deep dive into how Zettelgarden evolved from basic vector search to a sophisticated entity processing system, exploring the challenges and opportunities in augmenting human knowledge connections
Zettelgarden: Building Your Intelligent External Memory
Adding conversational AI to your personal knowledge management system
Retrieval Augmented Generation in Practice: Building Search for Connected Notes
Chunking, reranking, and vector search implementation with LLMs in Zettelgarden
Zettelgarden and Retrieval-Augmented Generation
Building retrieval-augmented generation into a personal information management system