Part VI: Shipping, Scaling, and Operating the Product
Chapter 28

Team Topologies and AI-Native Operating Models

The best AI PM you know just left to join a startup. She took with her knowledge of prompt engineering, eval design, and model selection that took two years to develop. Your organization has no mechanism to capture that knowledge, no role structure that would let someone else develop it, and no eval-driven workflows that would reduce the dependency on any single person. Traditional team structures were designed for deterministic software where knowledge can be documented and processes followed. AI products require new team topologies that treat institutional learning as a core competitive advantage.
The Tripartite Loop in Team Topologies

Organizing AI teams requires all three disciplines: AI PM defines roles, responsibilities, and how different functions collaborate; Vibe-Coding tests different team structures to find what actually improves delivery; AI Engineering implements the processes, hand-offs, and documentation that make team structure work.

Chapter 28 opener illustration
Team topologies define how human teams work with AI systems.
Vibe-Coding in Team Workflow Exploration

Vibe coding enables teams to explore and test different collaboration patterns before committing to new structures. Prototype how PMs, designers, and engineers would work together with vibe coding practices, test eval-driven workflows, and experience how different team topologies affect iteration speed. Vibe coding team workflows helps organizations discover the collaboration patterns that actually work for their context rather than adopting generic frameworks that may not fit.

Objective: Design team structures and collaboration patterns that support AI-native product development.

Chapter Overview

This chapter covers team topologies for AI products, including cross-functional structures, platform teams, central AI enablement, collaboration patterns, and AI review boards. Includes case studies from HealthMetrics and DataForge, plus org chart templates for different maturity stages.

Four Questions This Chapter Answers

  1. What are we trying to learn? How to organize teams to support AI-native product development with appropriate collaboration patterns and clear responsibilities.
  2. What is the fastest prototype that could teach it? Mapping your current team structure against AI-native collaboration requirements to identify gaps and barriers.
  3. What would count as success or failure? Team structures that enable rapid AI iteration without sacrificing quality, governance, or collaboration.
  4. What engineering consequence follows from the result? Traditional team structures often impede AI product development; topology choices directly impact AI product success.

Learning Objectives

Sections in This Chapter

Cross-References