Introduction

The argument is in the title. Mastery is dead.

Strip away the rhetoric and the claim has three layers. I want to lay them out at the start, because the rest of the book is going to assume you have these in your head.

The first layer is the obvious one. The traditional path to mastery, the ten or fifteen years of accumulated craft that used to define an expert in any field, is no longer producing the same return on investment. Not because the expertise is wrong. Not because the people who have it are unskilled. The technology around them has changed in a way that compresses what their expertise produces. The accountant who spent two decades getting fast at audit work is now competing with a tool that does audit work in minutes. The lawyer who spent fifteen years getting fast at contract review is now competing with a tool that reviews contracts in seconds. The graphic designer who spent ten years mastering execution is now competing with a tool that executes at near-professional quality on demand. The expertise has not disappeared. The market for it has, or is going.

The second layer is the more honest one. Mastery as we knew it was always two things stacked on top of each other. There was the production layer, the part where the expert actually produces the artifact: the audit, the contract, the design, the spec, the analysis. And there was the judgment layer, the part where the expert decides what to produce, why it matters, who it is for, and what to do when the unexpected happens. For most of the history of expert work, these two layers were sold as a bundle. The production was visible, the judgment was implicit, and the market priced the bundle as a single thing. AI compresses the production layer. It does not compress the judgment layer, at least not yet. Which means the value of the bundle is being unbundled in real time. The professionals who built their identities around production are watching their value collapse. The professionals who built their identities around judgment are watching their value rise, sometimes dramatically. Most professionals were doing both, and the unbundling is forcing them to figure out, often for the first time, which side of the line they actually live on.

The third layer is the strategic one. Mastery used to be a moat. You spent ten years getting good at something, you accumulated experience nobody else had, and that accumulation protected you. The moat was the thing that justified the salary. It was the thing that made you hard to replace.

The moat is gone, or going. Not because expertise is worthless but because expertise is no longer scarce in the same way. A motivated generalist with the right AI stack can now do work that previously required a specialist with a decade of training. The specialist's depth has not stopped mattering, but it has stopped being defensible. The new moat is not depth. The new moat is the ability to construct your own leverage. To assemble a personal apparatus of tools, judgment, and taste that lets you operate at the surface area of an entire team. The new mastery is not a body of knowledge. It is the willingness to build, configure, and maintain your own machine.

These are the three layers. Production mastery is dying. Judgment mastery is rising. The new mastery is leverage construction. The book argues these three things across twenty-five different careers, with enough specific evidence that by the end the reader can see the same pattern playing out everywhere they look.

I should tell you what I think is actually happening, because most of the writing on this topic has it half wrong.

The dominant narrative right now is that AI is going to take everyone's job. This is the doom version. It is wrong, or at least it is wrong in a way that matters. The careers in this book are not being eliminated. They are being restructured. The work that constitutes them is shifting. The number of people each career can support is shrinking, sometimes by a lot. The path into the career is closing for many of the people who would have entered it five years ago. But the careers themselves persist in altered form, and the people who navigate the change well are coming out the other side better than they went in.

The other dominant narrative is that AI will free everyone to do creative, fulfilling work while the boring stuff gets automated. This is the optimist version. It is also wrong, in a different way. The work that is being automated is not the boring work. It is the load-bearing work that justified the salary. The PRD, the contract review, the audit, the spec, the brief. These were not boring tasks. They were the visible labor that made the expert's value legible to the client. When that work disappears, the expert is left with the parts of the job that were always implicit, always discounted, always assumed to be done in the margins. Those parts can absolutely become more valuable. They can also become harder to charge for, because they were never priced as a separate thing.

The truer story, the one this book is trying to tell, is something like this. AI is compressing the production layer of expert work across an enormous range of careers, faster than the institutions that train and employ those experts can adapt. The compression is uneven. Some careers are getting hollowed out almost completely. Others are evolving in ways that will leave most current practitioners behind. A few are coming out the other side stronger and more interesting than they were before. The professionals who win the transition are the ones who see the compression early, accept that it is real, and rebuild their work around the parts of the job AI cannot do. The professionals who lose the transition are the ones who keep optimizing the production work AI now does for free, hoping the floor will hold.

The floor will not hold. Pretending it will is the most expensive mistake a professional can make right now.

A few years ago I started running my businesses differently. I have a digital design agency. I have a Shopify boutique my wife and I built together. I have a couple of other things that are still finding their shape. For most of my career as a builder, running these businesses meant hiring people to do the work I could not do alone. A designer here, an analyst there, an ops person to keep the wheels on. The math was simple: revenue minus payroll minus everything else equals what I take home. Growth meant hiring more people to do more work.

Around the start of 2024 the math started to change. I noticed that work I used to delegate to people I had hired could be done faster and, in some cases, better by AI tools I had started using on the side. At first this was a productivity story. I could ship more work in less time. Then it became a margin story. The hours I was paying employees for were starting to look like hours that did not need to be paid for, because the same output was coming out of a Claude subscription and an evening of focused work.

Then it became something harder. Some of the people I had hired were doing work that no longer needed a person to do it. Not all of their work. But enough of it that the role I had hired them into stopped existing in the form I had hired them into. I had to choose, role by role, whether to expand the role to absorb the freed time, or whether to let the role go.

I did not let everyone go. Some of those roles expanded. The person doing them moved up into work that was harder to automate, and the company got better. But I let some of them go. The conversations were not abstract. They were specific people with specific lives, and I was telling them that the role I had brought them into no longer existed.

I am telling you this for one reason. Most of the writing about AI and work is being done by people who have not had to make those decisions. By journalists watching from the outside. By academics theorizing about the future. By career coaches selling courses about navigating change. The people who have actually run small businesses through this transition, who have hired and fired during it, who have looked at someone they liked and decided whether the role they hired them for still existed, those people have a different read on what is happening. The view from inside is not the view from outside.

The view from inside is that this is happening faster than the discourse admits, that the changes are more fundamental than the discourse admits, and that the people who think they have time to figure it out probably do not. I am writing this book because I think that view deserves to be in print, and because I think a lot of the people reading it are going to need to make similar decisions about their own roles within the next few years.

The book has a structure worth understanding before you start.

It is twenty-five chapters, each focused on a specific career. The careers are organized into five parts. Part One is the Code and Numbers People. The professions where AI compression has been most aggressive and where the evidence is overwhelming. Software engineers, data analysts, financial analysts, accountants, actuaries. Part Two is the Word People. The professions where language is the product. Copywriters, content marketers, journalists, translators, technical writers. Part Three is the Visual People. Where execution is collapsing fastest and taste is becoming the only durable asset. Graphic designers, product designers, photographers, video editors, illustrators. Part Four is the Builders of Companies. The connective-tissue careers, the ones built on translation and integration and coordination. Product managers, project managers, consultants, marketing managers, operations managers, recruiters. Part Five is the Advisors and Gatekeepers. The careers that exist because knowledge used to be hard to access, and what happens to them when the asymmetry collapses. Lawyers, real estate agents, HR, insurance underwriters.

After the twenty-five careers, the book pivots. Five chapters in a closing section step back from individual careers and look at the larger pattern. What we have learned about expertise itself across all twenty-five fields. Who wins this transition and who loses, with the demographic patterns spelled out honestly. What new businesses become possible when expertise compresses into tools. What kinds of mastery survive and what they look like in their new form. And finally, a letter from me to you, written directly, with whatever the book has earned by the time we get there.

Each career chapter follows the same structure, so you will start to recognize the rhythm by the third or fourth one. There is a framing section that opens with how the role looks from a particular angle. A history section that includes the historical analog for the current compression. A specific-evidence section about what AI is doing inside the role right now, with named tools and concrete examples, and in some chapters a passage from my own direct experience. A residue section about what cannot yet be compressed. A playbook section with principles for the practitioner. A "What I Would Do" section where I tell you, in first person, what move I would make if I were in your role tomorrow morning. An honesty section about what is being lost. And a forward-leaning section about what new ground the compression opens up.

The book is designed to be read several ways. Cover to cover is the most rewarding read. Jumping to your career first and reading the closing section after is the most efficient. Sampling chapters from different parts to see the shape of the disruption is what I would recommend if you are a leader thinking about how to restructure an organization. There is a short section called How to Read This Book right after this introduction that walks through this in more detail.

A few things this book is not.

It is not a celebration of AI. I have spent enough time inside these tools to know what they can and cannot do, and I have no patience for the version of the AI conversation that promises a frictionless future. The transition is going to be hard for a lot of people. Some of those people deserve better than they are going to get. The market is not going to be careful with them. The book takes that seriously and does not paper over it.

It is not a doom book either. I am not going to convince you that AI is going to take everything from everyone. The honest read of the situation is that some careers are getting hollowed out faster than anyone admits, some are evolving in ways that will leave most current practitioners behind, and some are going to come out the other side better than they went in. The twenty-five chapters are an attempt to look at twenty-five specific careers and tell you, as honestly as I can, which is which.

It is also not a prediction book. I am not in the business of telling you what the world looks like in 2035. The technology is moving too fast and the second-order effects are too tangled to make confident long-range predictions, and anyone who tells you they can is selling you something. What I can do is tell you what is happening right now, what is likely to happen in the next two to five years given current trajectories, and what to do about it. The book is built on a near-term horizon because that is the horizon a working professional actually needs to plan against.

It is also not a book about what to do if you are a senior executive deciding how to deploy AI in your company. There are a hundred of those books and they are mostly fine. This is a book about what to do if you are the person whose job the AI is changing. It is written for you, not for the person making the decision about your role. The two perspectives are different and the book is honest about which one it is taking.

Six recurring threads run through the book. I want to name them now, because they will surface in different forms in nearly every chapter, and naming them upfront makes the book easier to read.

The first is build, don't buy. The professionals who win this transition are the ones who construct their own systems rather than waiting for vendors to package them. The reader who waits for an HR-approved AI tool from their employer is the reader who is already losing. The reader who has stitched together their own stack is the one pulling ahead.

The second is the compression itself. Expertise that used to take years to acquire is now embedded in tools that anyone can use after an afternoon of practice. Each chapter shows the compression in its specific form, and by the end of the book the argument is irrefutable not by force of rhetoric but by sheer accumulation.

The third is the widening gap between the people who use AI as an oracle and the people who use it as leverage. The oracle user asks the AI for an answer and accepts what it gives. The leverage user uses the AI to multiply their own thinking. The gap between them is widening fast, and the gap is between people, not just between practices.

The fourth is the return of the generalist. For two decades, work has specialized. AI is reversing this. The role of the future is not the deeply specialized expert. It is the generalist who can orchestrate a swarm of AI tools to produce work across multiple domains.

The fifth is grief. People are grieving the loss of their craft, their identity, the path they thought they were on. The discourse about AI does not have room for this grief. The book does. Every chapter has a moment where it pauses to name what is being lost.

The sixth is what becomes buildable. Every compression of expertise into tools opens up new businesses, new roles, new opportunities. The book is not just about what is being lost. It is also about what is becoming possible.

There are two more threads, one personal and one philosophical, that you will see less explicitly but that shape how the book is written. The personal one is the perspective of someone who has been around hard transitions before, in the Marines and in business, and who treats this transition with the seriousness it deserves without panicking and without minimizing. The philosophical one is the redefinition of mastery itself. The title says mastery is dead. The book proves the title is half true and half wrong. The mastery that is dead is the production-craft version. The mastery that is rising is the judgment-and-taste version, and it is harder, rarer, and more valuable than the old one ever was.

By the time you finish the book you will have seen these threads play out across twenty-five different fields, and the threads will have woven into a single argument about what is happening to expert work and what to do about it.

Here is who this book is for.

If you are early in your career and trying to figure out where to put your energy, the book is for you. The advice is going to be uncomfortable in places, because the honest read of some of these career paths is that they are not what they used to be and may not be worth a decade of climbing. I would rather tell you the truth and let you decide than feed you the conventional wisdom and watch you waste a decade.

If you are mid-career and noticing the ground shifting under you, the book is for you. You probably already feel some version of what the book describes. The book gives you names for what you are feeling and a path through it.

If you are senior and watching the people below you get compressed in ways you cannot fully explain, the book is for you. Some of what the book says about senior practitioners will flatter you and some will sting. Both halves are true.

If you are a leader trying to figure out how to restructure an organization, the chapters in your part of the world are useful. The closing section is more useful.

If you are not in any of these categories and just want to understand what is happening with work right now, the book is for you too. You are going to come away with a more honest picture than the one you are getting from the news.

There is one group I want to name specifically because the book is going to be especially useful to them. If you are reading this and quietly wondering whether to leave a corporate role and start something on your own, the book is in part written for you. The compression of expertise into tools is not just a threat to careers. It is also the largest opportunity to build a small, profitable, AI-augmented company that any of us has seen in our lifetimes. The closing section gets specific about what becomes buildable and how. If you are circling that decision, read the career chapter for your current role, then jump to chapter 28. The two together will tell you most of what you need to know.

One last thing before we start.

This book exists because I think the conversation about AI and work is mostly being had at the wrong altitude. The takes are too high, too abstract, too dependent on predictions about what the world will look like in ten years. The professionals reading this do not need predictions about ten years from now. They need a clear-eyed view of what is happening this year, in their specific career, and a credible playbook for what to do about it. The book is my attempt to give them that. I am writing it as a builder, not a journalist. The view from the builder's chair is different from the view from the press box. The builder is the one making the calls about which roles still exist in his company and which ones do not. The builder is the one who has had to look at someone he hired and tell them their role has changed past the point of recognition. The builder is also the one who has watched, in real time, what becomes possible when the production layer of expert work compresses into a tool stack. That perspective is the one I have, and the book is written from it.

I have tried to be honest about what I do not know. I am not the kind of expert who has spent thirty years inside any single one of these careers. What I have is the eye of someone who has worked across a lot of fields, has used these tools intensely, and has watched the patterns repeat. The book leans on that eye. Where I can vouch for something from direct experience I do, plainly. Where I am drawing on observation rather than experience I tell you. Where I am uncertain I tell you that too.

The next piece of the book, before the career chapters begin, is a short section called How to Read This Book. It walks you through the chapter structure and helps you decide which order to read the chapters in. After that, Part One opens with the Code and Numbers People. Software engineers, the canary in the coal mine, are chapter one. We start there because it is the career where the compression is most advanced, the evidence is most overwhelming, and the implications for everyone else are most visible.

Mastery is dead. Mastery has also, in a different sense, never been more valuable. The rest of the book is going to make those two sentences mean what they need to mean.

Let's go.