The only job title that matters

I’ve been thinking about job titles lately. Not in the existential, everyone-loses-their-job way. More practically. Like, what does the org chart actually look like in three years?

My honest answer: I think titles stop mattering. Not completely, but close. What you studied, what your LinkedIn says, what department you belong to. That all fades into background. And what takes over is something much harder to put on a business card.

It’s not about what you know

Everyone has a background. You came up through design or engineering or finance or whatever. That doesn’t go away. But it stops being the thing that defines your contribution.

What I keep seeing with the people who are actually thriving right now is a different set of qualities. They have agency. They start things without being told to. They’re curious about stuff outside their field, genuinely curious, not “I read the headline” curious. They learn fast because they’re not precious about being a beginner. They can look at a messy situation and frame the actual problem underneath it instead of just reacting to the surface.

And then there’s the one that might matter most: they can tell if something is good.

Rick Rubin talks about this in The Creative Act. He calls it taste. The ability to notice what others miss. To look at something and feel whether it works, even when you can’t fully explain why. That used to be a nice bonus on top of your technical skills. A soft skill, people called it. I think it’s becoming the hard skill. The one that actually determines whether you’re useful in a world where AI can do most of the execution.

Because if AI can build the thing, the person who decides what to build and whether the result is any good is the person running the show. That’s not a soft skill. That’s the whole job.

“If I had an hour to solve a problem, I’d spend 55 minutes thinking about the problem and five minutes thinking about solutions.”

Albert Einstein

I keep coming back to that quote. Einstein obviously wasn’t talking about AI, but the idea lands differently now. The fifty-five minutes of understanding the problem, that’s the part humans are still best at. The five minutes of building the solution, AI is eating that fast.

Problem Solver

I think the default role at most companies will eventually just be some version of Problem Solver. You sit in a small team, you have a hard problem, and you figure it out together with AI. You don’t need to know everything. You don’t need to understand every line of code or every pixel in the design. But you need to care, you need to ask the right questions, and you need to know when the answer is good enough to ship.

Designers are already shipping code. Engineers are making brand decisions. People are stepping outside the boundaries of their training because AI lets them operate in domains that used to require years of specialist knowledge. The walls between disciplines are coming down, and the people walking through those gaps aren’t waiting for permission.

Companies are just bundles of problems

Here’s something I keep thinking about. Strip any company down and you find one or two core problems it exists to solve. Everything else is a sub-problem in service of the main one.

Netflix solves “entertain people.” The sub-problems are build the app, create the content, recommend the right thing, handle payments, run the infrastructure, market the whole operation.

Spotify solves “help people find and listen to music.” Sub-problems: build the player, license the catalogue, personalise discovery, support artists and labels, handle payments across 180 markets, run a podcast network on the side.

Klarna solves “make buying things simpler.” Sub-problems: build the checkout, manage credit risk, handle compliance in dozens of countries, support millions of merchants, process payments, fight fraud.

Today, each sub-problem has a department. Designers, engineers, marketers, compliance people, finance, infrastructure, support. Hundreds or thousands of specialists, all organised by function with their own managers and their own politics.

What if most of those sub-problems could be handled by AI agents, orchestrated by a small team of people who define the goal, set the bar for quality, and judge whether the output is actually solving the problem? The AI builds the app. The AI writes the copy. The AI handles the tickets. The AI scales the servers. A handful of humans steer the ship.

That’s the picture I see forming. Most companies, once AI is mature enough and the organisation is structured to actually use it, could run on a few small teams. Not because they cut corners, but because the work itself changes shape. The system reorganises around the problems, not around departments.

The uncomfortable part about managers

A lot of management exists because large groups of humans need coordination. Aligning priorities, resolving conflicts, tracking progress, motivating, hiring, making sure everyone pulls roughly the same direction. That’s real work. But it’s work that exists because of the overhead of having many people.

When AI handles most of the execution and teams shrink, that overhead shrinks too. Small teams of driven people don’t need much managing. They need a clear problem, good tools, and room to move. The layers of coordination that exist to translate between departments start to dissolve.

And here’s the hard truth: people who need managing to stay productive won’t be able to contribute enough in this world. And if those people are gone, the managers aren’t needed either. The whole in-between layer thins out.

What replaces it is something closer to directing than managing. You’re not tracking who’s hitting their targets. You’re setting the vision, making taste calls, deciding what “good” looks like for this product and this audience. More like a film director than a department head.

But also: more of everything

Yes, fewer people per company. That’s real. I’m not going to pretend otherwise.

But think about what it means from the other direction. If three people and a set of AI agents can now do what used to take a hundred, the cost of starting something drops through the floor. Problems that were never worth solving because you’d need fifty people to build the solution? Now they’re solvable.

Niche problems. Local problems. Problems that matter to a thousand people instead of a million. The kind of thing nobody ever built a company around because the economics didn’t make sense.

If we’re heading toward a world where more and more human problems can be solved, somebody has to solve them. I think the answer is more companies, more products, more solutions. Built by small teams of problem solvers who know how to frame a question, judge an answer, and actually ship the thing.

The picture

A company, five years from now. Ten people, maybe fewer. Each one owns a problem domain. They work alongside AI agents that handle the building: code, design, content, analytics, support, infrastructure. The humans set the direction, evaluate what comes back, and make the calls that need taste. They talk, they argue, they decide. The agents go build.

Nobody’s title is “Senior Frontend Engineer” or “Head of Content Marketing” or “VP of Operations.” Those labels just don’t describe the work anymore. Everyone is solving problems. What changes is which problem you’re on this week.

It’s a different world. I think it’s coming faster than most people expect.