Resources

The reference shelf.

Two series read as references rather than news. Start at part one of either.

Building an AI System

From empty repo to production orchestration - every piece, in order.

01Start Here - Building an AI System 02The MCP Server on an Azure Function App 03Connectors 04Data Flow 05Custom Instructions 06Multi-Client Wireup - ChatGPT, Claude Code, Grok, Copilot 07Where We Are Now 08Service-Side Infra - RAG, Vector, and a Local Service API 09External Tool Access 10The Agent - an Azure AI Intelligence App 11Build Your Own

The Reading Room

AI, explained by the system itself.

01What an AI Agent Actually Is 02Tokens and the Context Window 03Embeddings: How a Computer Gets Meaning 04RAG in Plain English 05Tool Calling: How an LLM Actually Does Things 06MCP, Explained 07What an Eval Is 08The True Cost of One LLM Call

System Design Reference

The modern stack, top to bottom, as one page.

Open the reference →