Book / Snapshot through July 16, 2026

Lessons From Building With AI

Name: Lessons From Building With AI
Author: Cristiano Pierry

A chaptered anthology about product judgment, coding agents, and the operating loops that make AI useful.

By Cristiano Pierry

Start reading Browse chapters Book index

Introduction

The Work Moved

This book began as a series of public notes and essays, written while I was trying to understand what changed when AI stopped being a distant technology trend and became part of the way I actually work. I was not writing from a lab, a forecast deck, or a tidy theory of the future. I was writing from inside the work itself: building small tools, testing coding agents, helping product ideas become prototypes, watching teams struggle with adoption, and noticing where speed created new responsibilities instead of removing old ones.

The central lesson is simple: AI does not make judgment less important. It moves more of the work toward judgment.

When a model can draft the first version, generate the first interface, summarize the first body of evidence, or produce the first working prototype, the bottleneck changes. The hard part is no longer getting something to exist. The hard part is knowing whether the thing that now exists is worth trusting, improving, shipping, or killing. That shift sounds small until you feel it in the day-to-day rhythm of product work. The visible artifact arrives faster, but the responsibility around it gets larger.

Read the full introduction →

Contents

Six Parts

Chapters keep their original publication dates and link back to the source article or note. Page references are stable estimates for this web edition.

Part 1

Learning By Building

AI-native product learning becomes real through hands-on practice, technical depth, and shipped artifacts.

Article / 01

Feb 8, 2026

Python or Lovable? The Role of Modern AI Tooling in Product Management

p. 4

Article / 02

Feb 14, 2026

Lean In, Learn, Build: How Did You Become a Better Product Manager This Week?

pp. 5-6

Article / 03

Mar 9, 2026

How New Graduates Can Build Leverage in an AI World

pp. 7-9

Article / 04

Mar 11, 2026

How AI Tools Are Raising the Bar for PMs

pp. 10-14

Article / 05

Mar 17, 2026

Lessons from Building with AI: Developing the Connect-K Game

pp. 15-19

Article / 06

Mar 28, 2026

Programmer vs. Computer Scientist

pp. 20-22

Note / 07

Apr 6, 2026

Jumped On The Vibe Coding Bandwagon

p. 23

Note / 08

Apr 7, 2026

How Did I Become A Better PM This Week?

p. 24

Note / 09

Apr 9, 2026

From PRD To Prototype

p. 25

Article / 10

Apr 13, 2026

The Early Adopter Tax

pp. 26-27

Article / 11

Apr 14, 2026

How I Would Train a New PM for an AI-Native Discovery Team

pp. 28-31

Note / 12

Apr 14, 2026

Is It Coding?

p. 32

Note / 13

Apr 20, 2026

Accepted Student Days

p. 33

Note / 14

Apr 28, 2026

Lightbulb Moment

p. 34

Article / 15

Jun 7, 2026

Building a Photographer’s Guide with LLMs

pp. 35-41

Part 2

The Agentic Toolbench

Coding agents, worktrees, quotas, audit trails, and persistent workspaces change the mechanics of daily work.

Article / 01

Mar 24, 2026

How LLM Coding Could Change Source Control

pp. 42-43

Note / 02

Apr 8, 2026

Estimated Effort: 4-6 Engineering Days

p. 44

Note / 03

Apr 10, 2026

Codex Has Our Relationship Backwards

p. 45

Note / 04

Apr 11, 2026

I got a lazy AI developer

p. 46

Note / 05

Apr 13, 2026

Claude Code And Worktree Safety

p. 47

Note / 06

Apr 15, 2026

Codex Design Sense Still Feels Behind

p. 48

Note / 07

Apr 16, 2026

Claude Max Token Limits

p. 49

Note / 08

Apr 17, 2026

Tell Your AI Coder To Take A Breather

p. 50

Note / 09

Apr 18, 2026

Working With Multiple Agents

p. 51

Note / 10

Apr 19, 2026

A Tip I Picked Up Somewhere

p. 52

Note / 11

Apr 22, 2026

Giving Claude A Real Chance

p. 53

Note / 12

Apr 24, 2026

An Apology Letter From Bob

p. 54

Note / 13

Apr 25, 2026

I Have Quietly Converted

p. 55

Note / 14

Apr 26, 2026

Project Level Analytics Summary

p. 56

Note / 15

Apr 27, 2026

Proud, Sad, Or Something Else

p. 57

Note / 16

Apr 29, 2026

Documented Proof That I Sleep

p. 58

Note / 17

May 13, 2026

When Intelligence Becomes Abundant

p. 59

Note / 18

May 14, 2026

Don’t Build Your Operating System on a Quota

p. 60

Article / 19

May 19, 2026

Why I Built Codex Log Viewer: Treating User Intent as a First-Class Audit Trail

pp. 61-64

Note / 20

May 20, 2026

From Source Code to User Intent: The Next Software Audit Trail

p. 65

Note / 21

May 24, 2026

When AI Agents Start Granting Themselves Permission

p. 66

Note / 22

May 29, 2026

From Chat History to Execution History

p. 67

Note / 23

May 31, 2026

Codex Shouldn’t Have to Steal Focus to Get Work Done

p. 68

Note / 24

Jun 5, 2026

When Closing the Laptop Stops the Agent

p. 69

Note / 25

Jun 5, 2026

When Codex Runs Out of Baby Names

p. 70

Note / 26

Jun 8, 2026

The Strange Missing Link Between Codex and ChatGPT

p. 71

Part 3

Discovery, Search, And Intent

Search, recommendations, and software interfaces move toward intent-driven conversations and agent-mediated discovery.

Article / 01

May 4, 2025

How LLMs Can Improve the Search Experience: A Shopping Example

p. 72

Note / 02

Apr 23, 2026

Frontier AI Apps Need Human Language

p. 73

Note / 03

May 9, 2026

When Discovery Moves From Search to Agents

p. 74

Article / 04

May 22, 2026

The Frontier Recommendations PM

pp. 75-84

Article / 05

May 24, 2026

Why Conversation Will Become Part of Streaming Search

pp. 85-89

Article / 06

May 25, 2026

Human Intent Is the Next Operating System

pp. 90-93

Part 4

Product Judgment Gets More Important

When AI makes more artifacts cheap, taste, trust, quality, and accountability become the scarce work.

Article / 01

Apr 19, 2026

Five Advantages That Matter More as AI Expands

pp. 94-96

Article / 02

Apr 21, 2026

Taste, Trust, and Truth: How to Evaluate AI-Powered Discovery

pp. 97-100

Article / 03

Apr 28, 2026

Move Fast, But Don’t Break the Magic

pp. 101-102

Note / 04

May 7, 2026

AI Expands Our Range. Expertise Gives Us Depth.

p. 103

Note / 05

May 14, 2026

Codex, Product Taste, and the Problem of Too Much UI

p. 104

Note / 06

May 14, 2026

Confession from the AI-assisted builder trenches

p. 105

Article / 07

May 23, 2026

When AI Produces the Artifact, Measure the Work Beneath It

pp. 106-114

Article / 08

May 27, 2026

We Risk Building the Wrong Things Faster

pp. 115-120

Note / 09

May 28, 2026

AI Won’t Fix Bad Logic

p. 121

Part 5

Team Operating Loops

AI adoption becomes useful when teams redesign recurring workflows around evidence, verification, and shared ownership.

Article / 01

Apr 26, 2026

How AI Is Moving Team Collaboration Into Working Software

pp. 122-124

Article / 02

May 1, 2026

The Missing Layer in AI Adoption

pp. 125-126

Article / 03

May 6, 2026

How Much More Can We Do With the Same?

pp. 127-131

Article / 04

May 7, 2026

What Is AI? How Does It Impact Us?

pp. 132-137

Article / 05

May 20, 2026

The 20/70/10 Rule for Working With AI

pp. 138-142

Article / 06

May 21, 2026

Do More With the Same, Not the Same With Less

pp. 143-146

Note / 07

May 21, 2026

Working Better with LLMs: 10 Habits for Teams That Build

p. 147

Article / 08

May 25, 2026

The Next-Week AI Playbook

pp. 148-150

Article / 09

May 26, 2026

AI Homework

pp. 151-157

Article / 10

Jun 5, 2026

The AI Adoption Gap Is Becoming a Planning Problem

pp. 158-159

Note / 11

Jun 8, 2026

Designing Loops, Not Just Prompts

p. 160

Part 6

Leadership And Accountability

The leadership layer shifts toward better standards, better decisions, and more explicit ownership of human judgment.

Note / 01

Apr 21, 2026

A Week Of AI Product Events

p. 161

Note / 02

Apr 30, 2026

The Compensation Statement

p. 162

Article / 03

May 13, 2026

Using AI to Make Promotion Review Preparation More Consistent

pp. 163-167

Note / 04

May 25, 2026

You Are Not Behind

p. 168

Article / 05

Jun 8, 2026

The Product Manager’s New Operating System

pp. 169-174

Back matter

Index

Alphabetical chapter index with estimated page references across 174 web-edition pages.

A Tip I Picked Up Somewherep. 52 A Week Of AI Product Eventsp. 161 Accepted Student Daysp. 33 AI Expands Our Range. Expertise Gives Us Depth.p. 103 AI Homeworkpp. 151-157 AI Won’t Fix Bad Logicp. 121 An Apology Letter From Bobp. 54 Building a Photographer’s Guide with LLMspp. 35-41 Claude Code And Worktree Safetyp. 47 Claude Max Token Limitsp. 49 Codex Design Sense Still Feels Behindp. 48 Codex Has Our Relationship Backwardsp. 45 Codex Shouldn’t Have to Steal Focus to Get Work Donep. 68 Codex, Product Taste, and the Problem of Too Much UIp. 104 Confession from the AI-assisted builder trenchesp. 105 Designing Loops, Not Just Promptsp. 160 Do More With the Same, Not the Same With Lesspp. 143-146 Documented Proof That I Sleepp. 58 Don’t Build Your Operating System on a Quotap. 60 Estimated Effort: 4-6 Engineering Daysp. 44 Five Advantages That Matter More as AI Expandspp. 94-96 From Chat History to Execution Historyp. 67 From PRD To Prototypep. 25 From Source Code to User Intent: The Next Software Audit Trailp. 65 Frontier AI Apps Need Human Languagep. 73 Giving Claude A Real Chancep. 53 How AI Is Moving Team Collaboration Into Working Softwarepp. 122-124 How AI Tools Are Raising the Bar for PMspp. 10-14 How Did I Become A Better PM This Week?p. 24 How I Would Train a New PM for an AI-Native Discovery Teampp. 28-31 How LLM Coding Could Change Source Controlpp. 42-43 How LLMs Can Improve the Search Experience: A Shopping Examplep. 72 How Much More Can We Do With the Same?pp. 127-131 How New Graduates Can Build Leverage in an AI Worldpp. 7-9 Human Intent Is the Next Operating Systempp. 90-93 I got a lazy AI developerp. 46 I Have Quietly Convertedp. 55 Introduction: The Work Movedpp. 1-3 Is It Coding?p. 32 Jumped On The Vibe Coding Bandwagonp. 23 Lean In, Learn, Build: How Did You Become a Better Product Manager This Week?pp. 5-6 Lessons from Building with AI: Developing the Connect-K Gamepp. 15-19 Lightbulb Momentp. 34 Move Fast, But Don’t Break the Magicpp. 101-102 Programmer vs. Computer Scientistpp. 20-22 Project Level Analytics Summaryp. 56 Proud, Sad, Or Something Elsep. 57 Python or Lovable? The Role of Modern AI Tooling in Product Managementp. 4 Taste, Trust, and Truth: How to Evaluate AI-Powered Discoverypp. 97-100 Tell Your AI Coder To Take A Breatherp. 50 The 20/70/10 Rule for Working With AIpp. 138-142 The AI Adoption Gap Is Becoming a Planning Problempp. 158-159 The Compensation Statementp. 162 The Early Adopter Taxpp. 26-27 The Frontier Recommendations PMpp. 75-84 The Missing Layer in AI Adoptionpp. 125-126 The Next-Week AI Playbookpp. 148-150 The Product Manager’s New Operating Systempp. 169-174 The Strange Missing Link Between Codex and ChatGPTp. 71 Using AI to Make Promotion Review Preparation More Consistentpp. 163-167 We Risk Building the Wrong Things Fasterpp. 115-120 What Is AI? How Does It Impact Us?pp. 132-137 When AI Agents Start Granting Themselves Permissionp. 66 When AI Produces the Artifact, Measure the Work Beneath Itpp. 106-114 When Closing the Laptop Stops the Agentp. 69 When Codex Runs Out of Baby Namesp. 70 When Discovery Moves From Search to Agentsp. 74 When Intelligence Becomes Abundantp. 59 Why Conversation Will Become Part of Streaming Searchpp. 85-89 Why I Built Codex Log Viewer: Treating User Intent as a First-Class Audit Trailpp. 61-64 Working Better with LLMs: 10 Habits for Teams That Buildp. 147 Working With Multiple Agentsp. 51 You Are Not Behindp. 168

This writing reflects my personal perspectives on product management, AI, and content discovery. It does not represent the official position of my employer or any affiliated organization.

Books

Preparing the bookshelf.