The AI Plumber
Field Notes
The reference library for engineers who ship AI to production.
Featured
Context Debt
The Problem Nobody Names: Context Debt in AI Systems
Most AI systems don't fail because of bad models. They fail because nobody wrote down why the last decision was made.
Governance
Why Most AI Projects Fail Before They Start
70% of AI project failures trace back to governance gaps, not model performance. Here's what the data actually shows.
Agentic Systems
The 6 Pipes Required Before Any AI Agent Gets Write Access
Before any AI agent touches production data, six infrastructure layers must be in place. No exceptions.
Explore More
Stop re-litigating yesterday's decisions. Start documenting tomorrow's.