Skip to content
ADLC / The Book

A book by Asaf Saar

Product Managementin theAgentic Era

PM was always about what and why.
Now we can start doing the how.

$19.99 USD, one time. An online living edition: every future edition is included.

Book cover: Product Management in the Agentic Era, by Asaf Saar. A dark hardcover with the title in ivory and orange and an infinity loop connecting Intent, Build, Evidence and Decision around the letters PM.

AI agents are collapsing the distance between an idea and a working product.

Product managers can now research, prototype, build, test, analyze, and iterate in ways that previously depended on a sequence of handoffs.

This book shows how to take part in building without giving up what product management is for: judgment, accountability, and the right to stop.

What you get

A book you read and work in.

22 chapters in 4 parts, read online in a reader built for working product managers, plus a companion career guide.

  • Passage search

    Find the passage that answers a question and jump straight to it.

  • A private notebook

    Bookmark chapters and keep notes beside the passages they belong to.

  • A practice workspace

    Shape a product bet in twelve prompts and export the brief.

  • Dated editions

    See what changed and why, and read any earlier edition as it was released.

The online reader: a chapter with the contents list and reading tools.
The reader
The practice workspace: prompts for shaping a product bet beside a live brief.
The practice workspace

Included: the companion career guide

Finding your way into product.

A companion guide of about 5,500 words for anyone moving into product management. Start with the section closest to your background, then use the twelve-week plan as a practice structure. It is not a promise of a job within twelve weeks.

  • What a product manager actually does

    Problems before features, giving a team direction, and what a working week can contain.

  • Is product management right for you?

    What the work asks of you, preferences that may point to another role, and a small piece of real work to test the fit.

  • Your starting point is an asset

    Routes in from software development, QA, design and user research, sales, and outside technology.

  • Earning your first product role

    A small product investigation, a portfolio of decisions, interviews, and a twelve-week practice plan.

The premise

The old Product Management contract is breaking.

For years, Product Management was built around a simple division of responsibility:

PM owns the what and the why.
Engineering owns the how.

That division made sense when execution was expensive.

A product manager could define an opportunity, write requirements, align stakeholders, and then hand the work to specialists who designed, built, tested, deployed, and measured it.

Agents change that equation.

The distance between intent and execution is shrinking.

And when PMs can increasingly participate directly in the how, the role itself begins to change.

From a chain to a loop

Before

  1. Research
  2. Requirements
  3. Design
  4. Engineering
  5. QA
  6. Data
  7. PM

A chain of specialists. A sequence of handoffs. Feedback measured in days, weeks, or months.

Now

  1. Intent
  2. Build
  3. Evidence
  4. Decision

With the Product Manager at the center. A continuous loop.

Not an “AI for PMs” book

This is not a book about ChatGPT.

Or Claude.

Or Cursor.

Or whatever launches next Tuesday.

Those tools will change. The deeper shift will not.

This book is about what happens when intelligence and execution become available on demand.

When a PM can move from an idea to a working prototype. From a question to an analysis. From a customer problem to an experiment. From an assumption to evidence.

Without waiting for every traditional handoff in between.

The important question is no longer simply how AI makes Product Managers more productive.

The important question is what Product Management becomes when PMs can do things that were previously outside the role.

From coordinator to builder

The job is not disappearing.
Its boundaries are.

  • FromWriting requirements
    ToExpressing intent
  • FromManaging handoffs
    ToOrchestrating humans and agents
  • FromWaiting for prototypes
    ToBuilding prototypes
  • FromRequesting analysis
    ToExploring evidence directly
  • FromPeriodic discovery
    ToContinuous learning
  • FromManaging a backlog
    ToManaging an adaptive system
  • FromCoordinating execution
    ToParticipating in execution
  • FromPM as coordinator
    ToPM as builder

This does not mean every Product Manager becomes an engineer, designer, researcher, or data scientist.

The modern PM can go further before a handoff is required. Sometimes much further.

The Agentic PM loop

Intent. Build. Evidence. Decision.

The traditional product process is often described as a sequence. Agentic Product Management increasingly behaves like a loop.

  1. Intent

    What are we trying to change? What problem matters? What outcome are we seeking?

  2. Build

    What can we make now to learn? A prototype. A working feature. An internal tool. A simulation. An experiment.

  3. Evidence

    What happened? What do users do? What does the data tell us? What did we learn?

  4. Decision

    Continue. Change direction. Scale. Stop. Then return to Intent.

The faster we can move through the loop, the less Product Management is about managing artifacts and the more it becomes about managing learning.

Inside the book

Four parts. One loop.

The book follows the loop on its cover, from intent to a decision, with firsthand cases from building and leading.

  1. Part I

    Intent

    The problem, the product role, and the direction given to a builder.

    • The Agentic Development Life Cycle
    • PM, Not PO
    • A Story the Builder Can Follow
  2. Part II

    Build

    Into practice, from the first attempt to personal and employee agents.

    • Finding Your Way Into the Work
    • The Build Loop
    • From My Agent to Every Employee
  3. Part III

    Evidence

    Technical checks, user outcomes, commercial commitments, and the agentic working model.

    • The Independent Check
    • Measure the Work That Matters
    • From a Working Product to a Business
  4. Part IV

    Decision

    Investment, authority, organizational rhythm, leadership, and professional development.

    • Continue, Change, Stop
    • Authority and the Right to Stop
    • Giving People Room to Build

See all 22 chapters

After reading

What you will be able to do

  1. Write a build brief

    Give an agent a story it can follow: the user, the constraints, and what done means.

  2. Set an agent's authority

    Decide what an agent may read, do and send, and who can stop it.

  3. Evaluate a generated result

    Separate a passing check from what the work actually establishes.

  4. Make the investment call

    Continue, change or stop a bet, with the reason on record.

Not automated away

Judgment becomes more important, not less.

When execution gets cheaper, choosing what should be executed matters more.

When prototypes become cheap, deciding which ideas deserve attention matters more.

When analysis becomes abundant, asking the right question matters more.

When agents can act, accountability matters more.

AI does not remove the need for Product Management. It changes where Product Management creates value.

Execution becomes abundant.
Judgment remains scarce.

Who this book is for

Written for the people redesigning product work.

  • Product Managers

    Who can already feel the job changing and want to understand what comes next.

  • Product Leaders

    Who need to redesign how their teams work, what skills they hire for, and how Product interacts with Engineering, Design, and AI.

  • Builders Moving Into Product

    Who can suddenly create products without waiting for a traditional development organization.

  • Founders and Executives

    Who need to understand how AI changes the structure of product development itself.

  • Product Organizations

    Who want to move beyond “using AI tools” toward an operating model designed around agents.

Why I wrote this

A note from Asaf Saar

I've spent most of my career building products and leading product organizations.

For years, the boundaries were relatively clear. Product defined the problem. Design shaped the experience. Engineering built the solution. Data measured what happened.

Then agents began changing the way I worked.

I wasn't simply writing requirements faster.

I was researching.

Building.

Analyzing.

Testing.

Creating working software.

Exploring ideas that would previously have required multiple people and multiple handoffs before I could learn whether they were worth pursuing.

That raised a much bigger question for me:

If Product Managers can increasingly do the how, what becomes of Product Management?

This book is my attempt to answer that question.

Asaf Saar

About the author

Asaf Saar is EVP and Chief Product Officer at Mend.io. His thirty-year career runs from QA engineer at NetManage through engineering leadership at SAP, co-founding TenKod, and product leadership at Perfecto, Sauce Labs and Tricentis. Today he leads leaders in product management.

The book grows out of his own building: a Commodore 64 game made with AI, a review in which green checks on his Dofek project hid three real problems, and the step from a personal agent, TARS, to agents for every employee.

He created the operating model the book calls ADLC, the Agentic Development Life Cycle. Others use the term too; here it means one specific model of development responsibilities inside the product outcome loop.

Asaf on LinkedIn

The connection to ADLC

Product Management is one part of a larger shift.

ADLC asks:

What happens to software development when agents become execution units?

This book asks:

What happens to Product Management inside that new system?

They are two views of the same transformation.

ADLC

  1. Intent
  2. Generate
  3. Validate
  4. Deploy
  5. Observe

Govern applies throughout

ADLC describes the development responsibilities. They overlap rather than run as gates, and Observe can redirect any of them. It operates inside the product outcome loop.

Agentic Product Management

  1. Intent
  2. Build
  3. Evidence
  4. Decision

Product Management in the Agentic Era describes how the PM operates inside it.

Explore ADLC

Read before you buy

Look inside the book.

Read a free sample before deciding whether the book is for you. It opens right here in your browser.

No email gate required.

Read the Sample

Get the book

Online living edition

Product Management in the Agentic Era

By Asaf Saar

PM was always about what and why.
Now we can start doing the how.

Edition
Online living edition
Includes
Every chapter, the companion career guide, the reader tools and every future edition
Access
Read online, signed in with your purchase email
Format
Online edition. Read on phone, tablet and desktop. No PDF or EPUB of the full book; the free sample is a PDF.

$19.99 USD

Checkout is handled by Lemon Squeezy. Applicable taxes are calculated at checkout. After purchase, sign in with the email you bought with to read the full book. Already bought it? Sign in. Terms · Privacy · Support

Buying for a team?

Team packages and workshops for product organizations are arranged directly. Reach out on LinkedIn.

The next generation

Product Management was built for a world where PMs defined the work and others executed it.

That world is changing.

The next generation of Product Managers will still own the what and the why.

But they will increasingly participate in the how.

Get Product Management in the Agentic EraRead a Sample First