API vs MCP: When Should You Use Each?
Use the interface that matches who's reading — humans, software, or agents.
Notizen, geschrieben während der Arbeit.
Technische Posts, Produktentscheidungen und ehrliche Beobachtungen — geschrieben beim Bauen, nicht danach.
KI-Engineering
Wie sich KI-Funktionen verhalten, wenn echte Menschen sie nutzen — nicht in kontrollierten Demos.
How I Would Build an Agent-Readable Personal Hub From Scratch
Start with the data model, then expose the same truth through pages, APIs, MCP tools, and examples.
Your MCP Server Is Only as Good as Its Tool Descriptions
An MCP tool is not useful because it exists. It is useful when the agent understands when, why, and how to call it.
Why Local-First Time Tracking Protects Your Privacy
Time tracking data is behavior data, and behavior data deserves local-first defaults.
Building a Public API for My Personal Hub (and AI Agents)
If an AI system is going to summarize your work, give it structured context instead of making it scrape and guess.
AI Agent Architecture: Patterns That Actually Work in Production
Agents fail at the seams, not the steps.
Prompt Engineering for Production: Moving Beyond Demos
Prompts that work in demos fail in production because demos don't have variance.
AI builders: chase primitives, not model size
Models impress. Systems ship.