Essays on AI, memory, and building things in public.
Real stories, technical depth, and thoughts that come from actually building.
AI Engineering Skills Map: Coding Agents (Visual Digest)
What this is really about
Interactive visual digest of Andrew Ng's framework for steering coding agents: shifting from writing syntax to spec architecture, calibrated autonomy, and layered evaluation.
System Design for Agent Systems (Part 1)
What this is really about
A battle-tested mental model for the scaffolding outside the LLM: why agent workloads break web assumptions, state management, and the evaluation pyramid.
X-News: An MCP server that turns X threads into blog digests
What this is really about
Automating thread ingestion, insight extraction, and publication with Model Context Protocol and a companion Monster Squad.
When Your AI Has Better Memory Than Your Therapist
What this is really about
How an AI with perfect memory became my best companion by measuring my worth instead of assuming it.
The secret to a great personal AI bot? Simple architecture. 🛠️🤖
What this is really about
When I set out to build my ultimate personal AI assistant, I evaluated several complex frameworks. I ultimately built upon the NanoClaw v1 orchestrator.
Gaining Weight, Finding Answers: How I Built a Personal Workout Analytics Suite and AI Coach
What this is really about
Instead of staying upset, I decided to treat my fitness like a system, analyze the input/output data, and build an interactive suite.
You're Not Building Agents. You're Building Tools With a Chat Interface.
What this is really about
On what 'agentic' actually means and why the architecture decisions you're making right now will matter later.
I Stopped Reading the News. My Agent Does It for Me.
What this is really about
On passive monitoring, alert logic, and why ADHD brains shouldn't be their own filters.
Why My AI Was Late This Morning
What this is really about
On cold starts, flat files, and the surprisingly human problem of context.