The Minimum Viable Company

Since January I have built more than most teams accomplish in six months. No engineering team. No product manager. No head of design. Just tools, judgment, and a lot of hours. I used to run a multimillion-dollar company, and I know exactly what execution used to cost. What I am doing now should not be possible by the old math.

But it is. And that changes everything.

Five years ago, building a startup required people. A technical co-founder or at least an engineer. Someone for product. Someone for design. Legal, bookkeeping, financial modeling, customer communication. These were not luxuries. They were the price of admission. Company formation was expensive because execution was fragmented, and even a strong founder eventually hit the wall of his own time and skill.

That constraint is breaking, and most people are still pretending it is not.

Intelligence is no longer scarce. It is a utility. The founder with the right stack can now reach for capabilities that used to sit behind payroll: analysis, code, research, iteration. What used to require multiple smart people across multiple functions can now be pulled into one fast-moving operator with judgment. That should unsettle more people than it does. A lot of careers were built on access to expensive cognitive labor. That wall is coming down, and most of the market still has not adjusted to what that means.

This is not incremental. It is structural. The cost and timeline of company formation are shifting in real time, and investors still using old assumptions are going to keep overpricing the wrong inputs.

I saw the shift before I had clean language for it. An AI was my first real thought partner. At first I used it the way most people did: questions, sparring, thinking out loud. Then the conversations got sharper, the output got better, and it stopped feeling like a toy. A coding model let me build in the trenches. Not abstract help, not vibes. Real code, real problems, real stuck points that needed to break open. A research tool changed how fast I could learn, letting me ingest and organize at a speed that would have felt absurd a few years ago.

At some point I looked up and realized I was not using isolated tools anymore. I was running a stack that functions like an operating system for execution. Research, writing, code, product thinking, iteration. Each tool has a job. Together they multiply. What used to require two or three capable employees can now be compressed into one founder moving fast across functions.

Specialization is not dead. But the founder no longer has to wait for specialists to begin.

Last June, a friend of mine who had operated at the absolute top of the technology industry told me something I was not ready to hear.

I was in a rough spot. New city, uncertain footing, questioning everything. He called me and said: stop everything you are doing. Learn everything you can about AI. Because in two years the white-collar labor market is going to look nothing like it does today, and if you want to survive this, you need to know everything.

He was not being dramatic. He was being precise. I did not listen fast enough. I am listening now.

The founders who win this era are not waiting for that realization. They are already operating inside it.

The founder who wins now is not the one who knows one thing deeply and waits for a team to fill the gaps. It is the one who learns fast enough to get competent across functions, uses the stack to close the distance between idea and execution, and ships before the slow money figures out what changed. The edge is not raw intelligence. It is speed of learning, speed of iteration, and the willingness to keep moving without waiting for permission.

It is not a personality type. It is a practice. A shrinking number of people are doing it. The rest are watching.

There is a version of this argument that sounds threatening if you are a specialist. Fine. But expertise is not dead. The tax on starting is. The founder who used to need a CTO to get to a working prototype can now use a capable model as a technical thought partner, get real code written, and arrive at a proof of concept before the first investor meeting. That does not replace engineers. It removes delay.

There is also a fantasy version of this story where AI does the work while you supervise from a distance. That is not what building looks like either. What changes is dead time. Research compresses. First drafts appear on demand. Code gets written. The founder who stays locked in can turn the same hours into dramatically more usable output because the bottlenecks are disappearing one by one.

AI does not make you magical. It makes iteration brutally fast if you stay in the work and keep your standards high. The people who understand that are already lapping the field. The ones still waiting to feel ready are going to look up one day and realize the gap closed without them.

Here is where it gets real for investors.

Pre-seed economics are shifting faster than most term sheets reflect. Milestones that used to require $1M and a full team can now be hit with $100,000 to $250,000 in the hands of the right founder. That means more shots on goal, less early dilution, and a real chance to get to proof before getting to payroll.

It also means the evaluation criteria have to change. A solo founder without a CTO on the team slide used to be a red flag. In some rooms it still is. That is lazy pattern matching. The real question is whether this founder can learn fast, build lean, and ship before running out of runway. Right now, that is more predictive than almost anything else you can evaluate at the pre-seed stage.

In this era, the founder is the factory.

The next breakout companies will not start with bigger teams. They will start with one person who has learned to multiply his capacity, who treats the stack as infrastructure and iteration as the job, and who stays in the arena long enough to find what works while everyone else is still writing job descriptions.

This is not a trend. It is a repricing of execution itself.

Get ahead of it, or spend the next five years explaining why you did not.