Pillars — Architectural Foundations
Eight deep pillars covering the design patterns that make AI systems work in production. Each pillar is anchored to a course; the supporting cluster articles dive into specifics.
MCP Architecture for Builders
Model Context Protocol turned ad-hoc AI tool integration into a standard. Five design patterns, seven failure modes, and what the 2026 roadmap means for the architecture decisions you make today.
Read pillar →Personal AI Knowledge OS
Building a cognitive operating system for solo operators. Knowledge graphs, vector retrieval, prompt libraries, context management — the design patterns that compound.
AI Agents Architecture for Solo Operators
Agent orchestration patterns from reality, not tutorials. Decision trees for single vs multi-agent, memory persistence, failure modes, cost monitoring.
AI Workflow Automation Design Patterns
Workflow architecture without vendor lock-in. Principles that translate between Zapier, Make, n8n, and custom — when to use each.
GEO / AI Overview Optimization
How to engineer content for AI engine citation. Google AI Overviews, Perplexity sources, comparison page architecture. The $365M market opportunity.
Production RAG Patterns
RAG architecture decisions: when not to use RAG, trade-offs, anti-patterns. Chunking, hybrid retrieval, reranking, evaluation — what survives real load.
AI Tools Stack for Indie Builders
My actual AI stack as a solo architect. Claude Code, Cursor, Lovable, observability, hosting — opinionated, dated, specific.
AI Income & Operations for Technical Builders
Sustainable AI income architecture. Multi-stream, multi-locale, architect approach. The honest economics of an AI-augmented solo operation.