Skip to content

Part II: Building

Infrastructure, tools, agents, and architecture patterns for AI-first companies.


Overview

With foundations in place, this section gets practical. We cover the infrastructure needed for AI operations, how to build software with AI coding tools, agent architecture patterns, and the microservice approach that enables rapid scaling.

Chapters in This Part

Chapter 4: Infrastructure for AI-First Operations

The infrastructure stack for AI-first companies. Covers polyglot persistence, the AI Tool Gateway pattern with tiered access (safe/supervised/forbidden), and unified auth for humans and agents.

Chapter 5: Building with AI — Coding Tools, Agents, and Patterns

How to build software with AI as a collaborator. Claude Code, Cursor, Copilot decision frameworks, and the patterns that enable 5x productivity.

Chapter 6: Agent Architecture — Chat, Background, and Orchestration

Two types of agents, when to use each, and the Agent Hub pattern. Designing for both interactive and autonomous AI operations.

Chapter 7: The Microservice Pattern for AI-First Companies

Domain microsites with shared infrastructure. The architecture pattern that enabled 15 microsites in 3 months.


Key Concepts Applied

  • Tiered AI Access - Safe/supervised/forbidden permission model
  • AI Coding Maturity - Autocomplete → generation → agents → orchestration
  • Agent Architecture - Chat agents vs background agents
  • Domain Autonomy - Own your domain, share your foundation
  • Probabilistic AI - Designing for uncertainty in agent operations
  • Human AI Collaboration - Human-AI workflows in development

Reading Guide

For technical leaders: All chapters are essential—this is your implementation guide.

For product leaders: Focus on Chapter 6 (agents) and Chapter 7 (microservices) for architectural context.

For startup founders: Read all four—these chapters are about building fast with AI.


Book Contents