The Age of Builders and Doers
How a small group of operators, plugged into AI agents and modern tools, can tap into effectively unlimited scale.
I’ve spent a lot of time over the last few weeks thinking about what this year is going to look like — and more broadly, how companies are going to be built from here on out.
If you zoom out and ignore the headlines for a second, there’s a pretty simple pattern emerging underneath all the noise:
We’re entering the age of builders and doers.
The companies that win over the next decade won’t be the ones that stack up the biggest headcount, the most layers of management, or the fanciest org charts.
They’ll be the ones where a relatively small group of people operates with the impact of a much larger company because they know how to do three things:
See clearly what needs to exist.
Build it themselves (or with a tiny crew).
Use AI and automation as a multiplier, not a buzzword.
If you’re a founder, operator, or aspiring leader, this shift should feel both a little terrifying and extremely exciting.
Let me explain.
The Old Playbook: Headcount as Strategy
For a long time, the mental model for building a company looked something like this:
Raise money.
Use that money to hire people.
Promote your “best” individual contributors into managers.
Layer more managers on top of managers as you grow.
Build big departments around functions: Sales, Ops, Marketing, Product, etc.
If you were a founder, success often got measured (quietly) in headcount.
“I run a 40-person sales org.”
“We’ve got 25 people in operations.”
“We doubled the team this year.”
I ran into someone at the Bears game this weekend who’s running a company with hundreds of employees. The thought of building something that large, from a people perspective, was honestly daunting. My gut reaction was, “Oh man, my goal is to keep our team as small as possible and just make sure everyone has tools that 10–100x their impact.”
Headcount was a proxy for progress.
This model made sense when:
Tools were primitive. A lot of work was truly manual.
Data was hard to access. Answering basic questions required whole teams.
Internal systems were brittle. “Don’t touch that spreadsheet, it might break.”
The coordination cost of getting things done was high.
If you wanted to build a new function, the assumption was: “We need to staff it.”
You’d hire a manager, then a few ICs, then a manager over those managers, and so on. If you wanted to “be a leader,” that often meant getting out of the work and into managing the work.
Today, that model is breaking.
What’s Changed: The Human Possibility Frontier
When people talk about AI, they usually frame it as a question of replacement.
“Will it replace SDRs?”
“Will it replace copywriters?”
“Will it replace data analysts?”
I think that’s the wrong frame.
A more accurate question is: What does one motivated person now have the capacity to do when they’re plugged into tools like Notion Agents, Claude, ChatGPT, Nano Banana — and effectively have access to unlimited scale?
Because that’s what has changed most dramatically in the last 12–18 months.
A concrete example
In the last few weeks alone, I’ve used AI tools like Notion AI and Claude Code to:
Pull together a customer intelligence database that previously would’ve taken a team weeks.
Build and compare multiple pricing and compensation models.
Map customer and prospect locations to plan travel and coverage strategy.
Create analytical deep dives into coverage gaps in lanes, regions, or customer segments.
Set OKR baselines from messy historical data.
I didn’t magically become a full-stack engineer or a data scientist. I didn’t learn to write production-level code.
What changed is:
The tools are finally good enough.
The interfaces are finally usable.
And I stopped accepting “that would be nice to have, but we don’t have the bandwidth” as an answer.
All of the tasks above would’ve historically required a chain of events:
Scoping a project.
Getting on the product or data team’s roadmap.
Waiting for someone to write scripts, build dashboards, update data pipelines.
Now, a founder who is willing to get their hands dirty and learn a new workflow can accomplish a lot of this in a few focused sessions.
This is happening everywhere:
Sales leaders building their own custom reporting or qualification workflows.
Ops leaders automating onboarding, billing, and fulfillment steps without asking for engineering time.
Founders prototyping new internal tools or even product features themselves before a single ticket gets filed.
The frontier of “what’s humanly possible for one person” has moved.
And we’re still underestimating how far.
From Function Leaders to Function Builders
In the old world, you hired “function leaders” to run departments.
Their job was to:
Set strategy.
Define org charts.
Hire and manage teams.
Put processes in place and make sure people followed them.
In the new world, I think that role gets replaced by function builders.
What’s the difference?
A function builder is someone who:
Looks at a blank page and can design how a function should work.
Gets in the weeds to build the first version — systems, workflows, tooling, documentation.
Uses AI and automation as default ingredients, not optional add-ons.
Operates with a builder’s mindset even if their title says “Head of X” or “VP of Y.”
They might still manage people. They might still have a leadership title. But their primary value is not how many direct reports they have. It’s their ability to create and own a working system.
This is a very different skillset than “I can run a 50-person team.”
If you’re hiring, the questions start to change:
Not: “Have you managed a large team before?”
But: “Have you ever built a function from scratch and made it work?”
Not: “How many people did you oversee?”
But: “What systems did you design? What workflows did you implement? What did you personally build?”
And if you’re an individual contributor or mid-level manager trying to grow your career, your playbook should change too.
Instead of optimizing for, “How do I get more people reporting to me?” you want to optimize for, “How do I become the person who can build and run an entire function with a small team and the right tools?”
Operators With Agency
There’s a particular kind of person who thrives in this environment. I think of them as operators with agency.
You probably know a few of these people already. They’re the ones who:
Don’t wait to be told what to do.
See the gaps and quietly build the systems to close them.
Learn adjacent skills without making a big deal out of it.
Volunteer to own messy, unscoped problems and bring order to them.
If they run customer success, they’re not just responding to tickets — they’re designing the onboarding journey, automating repetitive touchpoints, and building health score systems.
If they run sales, they’re not just managing pipeline — they’re building a repeatable motion, instrumenting data, and automating the mind-numbing admin work.
If they run people ops, they’re not just scheduling reviews and onboarding new hires — they’re building the internal infrastructure that keeps the whole company running smoothly.
They’re builders disguised as “leaders” or “ICs.”
And when you pair people like this with modern tools — especially AI-native tools — they become unfair advantages.
Small Team, Big Impact
At Cargado, we’re deliberately not trying to build a 250-person company right now.
Our ambition is big. The market is massive. The problems we’re chasing are real and complex.
But structurally, we’re aiming for something different:
A team of fewer than 40 people that operates with the impact of 250.
That’s not a cute slogan. It’s a design principle.
What makes that feasible?
1. The right people
You can’t fake this with slogans on the wall.
You need people who are:
Comfortable owning outcomes, not just tasks.
Willing to learn new tools and workflows, even if it means being bad at them for a bit.
Motivated by building, not just maintaining.
Frustrated by unnecessary bureaucracy.
These are not always the people who look best on paper. A lot of them never had the “SVP” title. Many have punchy, nonlinear careers. But they’re the ones who can 10x your surface area when you put the right systems around them.
2. An AI-native mindset
There’s a big difference between:
“We use AI to write a few emails here and there,” and
“We’re building our company on an AI foundation.”
When I say “AI-native,” I mean:
You assume any repetitive or predictable workflow can be automated or assisted.
You design your data and systems so that agents can actually do high-quality work.
You expect people to use AI to extend their skillset, not replace their judgment.
We’ve built dozens of internal agents that do everything from enriching data to posting updates to coaching calls. They’re not perfect. A lot of them are v1s. But they move the floor up for what any one person can get done — especially when you pair them with tools like Claude, ChatGPT, and Nano Banana that give individual operators access to almost unlimited scale.
3. A bias for shipping tools, not just writing docs
Documentation is important. So is alignment. But in my experience, the fastest way to change how people work is to hand them a real tool that makes their life easier.
A script that turns a 45-minute manual task into a 3-minute check.
A workflow that automatically routes and enriches data instead of asking humans to do it.
A dashboard that surfaces the right questions so humans can focus on judgment, not retrieval.
When people can feel the time they’re getting back, they’re much more willing to lean into the change.
What This Means for Founders and Leaders
If you’re building or leading a company right now, I think there are a few uncomfortable but necessary questions to ask.
1. Are you hiring builders or administrators?
Look at your leadership team and your hiring pipeline.
How many people are there because they’re great at designing and building systems?
And how many are there because they’re good at managing existing ones?
You still need some of the latter, especially in regulated or complex environments. But if you skew too far in that direction, you end up with beautifully managed stagnation.
You want a bench of people who can:
Take a blank page and sketch the function.
Build the first draft of the system.
Iterate in the real world.
2. Where are you still solving problems with headcount?
When something isn’t working, the reflexive answer is often, “We need more people.”
Sometimes that’s true. More often, it’s a sign that you haven’t really redesigned the work.
Before you post the job description, ask:
Could a better system or workflow solve 80% of this?
Could an internal tool or agent handle the repetitive parts?
Is the real problem clarity, not capacity?
Headcount should be the last lever you pull, not the first.
3. Are you acting like AI is a side project?
A lot of companies have “AI initiatives” right now.
A tiger team.
A few experiments.
A slide in the board deck.
But if the core rhythm of the company hasn’t changed — if the way sales works, the way ops works, the way product works all look mostly the same — you probably haven’t internalized what’s possible yet.
You don’t need a 200-page AI strategy doc. You need to change how work actually gets done on Tuesday afternoon.
That usually starts small:
One workflow that goes from manual to automated.
One team that adopts agents as core tools.
One leader who starts building instead of waiting.
Once people see the step-function change in their own jobs, it becomes much easier to layer bigger changes on top.
What This Means for You (As an Individual)
Let’s zoom back down from “company strategy” to you — the person reading this.
If you strip away the noise, the opportunity is pretty wild:
A motivated individual who leans into this shift can operate at a level that used to require a full supporting cast.
You don’t have to wait for your company to “get its AI strategy together” to benefit from this. You can start with your own work.
Some practical starting points
Map your work into three buckets:
Pure judgment (shouldn’t be automated).
Pattern-driven but complex (great for AI assistance).
Repetitive, rules-based tasks (should be automated yesterday).
Pick one annoying repetitive task and automate it.
It might be:
Weekly reporting.
Cleaning up data from one system before loading into another.
Generating the first draft of emails or docs you write over and over.
Learn enough about the tools to be dangerous.
Not just chat interfaces. Learn:
How to connect them to your data.
How to call APIs or use no-code automation tools.
How to create simple agents that work on your behalf.
Redesign one small “function” in your world.
That might be:
How your team onboards new customers.
How you qualify and route inbound leads.
How you collect, store, and use customer feedback.
Treat it like a product. Design it. Build v1. Ship it. Iterate.
You’ll learn more about your own value and potential from that one project than from another year of status meetings.
The Emotional Side: Letting Go of Old Status Markers
There’s a quieter, more emotional dimension to all of this that’s worth naming.
For a long time, we’ve associated “success” with:
Bigger teams.
Fancier titles.
More direct reports.
Owning a big chunk of the org chart.
When you start talking about small teams doing more with AI and automation, it can feel threatening. “If the company doesn’t need a 40-person team, where does that leave me?”
I get that.
I’ve had to unlearn some of those status markers myself. I’ve also seen what happens when leaders cling to the old markers and resist the new reality.
Here’s how I’m trying to frame it for myself:
Status based on org chart size is fragile. It’s dependent on budgets, markets, and investor sentiment.
Status based on your ability to build and operate systems that work is much more durable.
If you become the person who can:
Take a fuzzy problem and turn it into a working function.
Blend human judgment with AI and automation effectively.
Increase the surface area and speed of your team without burning them out.
You will not lack for opportunity.
It might look different than the old “rise through the ranks and get a VP title at 32” story. But it will be more resilient, more interesting, and, in my opinion, more fun.
Where We Go From Here
I don’t think AI is a temporary productivity blip. I think it’s one of the most consequential technologies the world has ever seen.
Not because it will replace everyone overnight, but because it fundamentally changes what small, highly aligned groups of people can do.
The companies that treat it as a side project will get some efficiency gains and maybe a nice case study.
The companies that treat it as the foundation for how they operate will:
Ship faster.
Learn faster.
Serve customers better.
Require less bureaucracy to coordinate.
They’ll look “smaller” on LinkedIn and much bigger in every way that matters.
This year, my personal focus — and our focus at Cargado — is to lean all the way into this reality:
To keep hiring builders and operators with agency.
To keep simplifying instead of adding layers.
To keep redesigning work so that small teams can do what used to take an army.
If any of this resonates, here are two questions I’d leave you with:
Where are you still defaulting to headcount as the answer?
What function in your world could you rebuild from scratch today, knowing what’s now possible with AI and automation?
Because in the age of builders and doers, the real competitive advantage isn’t how many people you have.
It’s how much a small number of the right people can get done when they’re connected to tools that give them effectively unlimited scale.