The Two-Person, $1.8 Billion Company

What Happened

Matthew Gallagher launched Medvi, a telehealth company selling GLP-1 weight loss medications, from his apartment in Los Angeles in September 2024. He spent $20,000. He used ChatGPT to write the platform code and customer service scripts. He used Claude for strategy and additional coding. He used Midjourney and Runway to generate ad creative.

The company hit $401 million in sales in its first full year. It now serves 250,000 customers and is tracking toward $1.8 billion in 2026 revenue. The entire operation runs with two people. Gallagher and his brother Elliot.

For context: Hims & Hers, one of the largest telehealth companies in the country, reported $2.35 billion in revenue last year. They have 2,442 employees. Their net profit margin is 5.5%. Gallagher's margin is 16.2%. Three times higher. With one-twelve-hundredth the headcount.

How he did it matters more than the headline number. Gallagher did not build a hospital. He did not hire doctors. He did not set up a pharmacy. He outsourced every regulated function to existing infrastructure providers. CareValidate handled the doctor network and prescription compliance. OpenLoop Health handled pharmacy fulfillment, shipping, and patient management.

Gallagher kept one thing: the customer. Branding, website, paid media, checkout flow, and service. Everything the customer sees and touches. That was his. AI built it. He directed it.

The FDA issued a warning letter to Medvi in February 2026 for misleading marketing claims about compounded GLP-1 products. Thirty telehealth companies received similar warnings. This is real. The story is not a fairy tale. But the structural lesson holds regardless of how Medvi's regulatory situation resolves.

Why It Matters

The standard response to this story will be: "Good for him, but I am not starting a telehealth company."

Fair. But that misses the point.

Gallagher did not build new technology. He assembled existing pieces in a new configuration. AI wrote the code. AI generated the ads. AI handled the customer service. Existing companies handled the regulated infrastructure. His contribution was the architecture. Knowing what to own, what to outsource, and how to connect the pieces.

That is the same set of decisions every director, VP, and business owner makes every quarter. The difference is that AI just changed the math on how many people you need to execute those decisions.

AI-literate individuals and small teams will produce disproportionate results compared to larger teams that have not figured out how to integrate these tools into their workflows. The question for someone running a team of 50 is not "will someone build a billion-dollar company with two people?" That already happened. The question is what that team of 50 accomplishes when 10 of them operate at this level of AI fluency.

Lessons for Leaders

The tools are commodity. The architecture is the advantage.

Gallagher used ChatGPT, Claude, Midjourney, and Runway. These are consumer-grade tools. Millions of people have access to them right now. The billion-dollar insight was not "use AI." It was "own the customer, outsource the compliance, let AI handle the execution." If your AI strategy starts and ends with "buy a tool," you are solving the wrong problem.

Revenue per person is the metric that matters now.

The Medvi story forces an uncomfortable conversation about headcount efficiency. If AI-enabled teams can produce 10x, 50x, or 200x the output of traditional teams, the question every executive will face is: do we need this many people doing this particular task? The companies asking that question with data will outperform the ones avoiding it.

The FDA warning is part of the story. Read it.

Medvi received a formal FDA warning for misleading marketing claims. Thirty telehealth companies got similar letters. AI can build a company fast. It cannot exempt that company from regulation. Speed without governance is a risk. The professionals who understand both the power and the constraints are the ones their organizations need most.

One More Thing

I have been thinking about this story all week for a specific reason.

Two years ago, I built 45 deliverables in three weeks. One founder. One AI assistant. The market value of what we produced was somewhere between $750,000 and $1.1 million. I did not fully believe the implications of what that meant. Not really. It still felt like an experiment.

Gallagher's numbers made it real for me. Not because $1.8 billion is relatable. It is not. But because the tools he used are the same ones I use every day. And the approach, owning the strategic layer while letting AI handle the execution, is the same approach I teach.

The gap between "I know AI exists" and "I have restructured how I work around it" is where most professionals sit right now. The difference is not intelligence or access. It is a decision. A decision to stop treating AI as a tool you open sometimes and start treating it as an operating layer that runs underneath everything you do.

That decision does not require $20,000 or a startup idea. It requires one honest audit of how you spend your time and one honest answer about which of those tasks AI could do better. If you want to build AI systems for your industry with people who understand your workflows, that is what AI Crew Call is built for.

"The billion-dollar gap between knowing AI exists and restructuring your work around it is not intelligence. It is a decision."

If someone handed you the same tools Gallagher used and said "you have 60 days," what would you build? And what is stopping you from starting that right now?

Harrison Painter

AI Business Strategist. Founder, LaunchReady.ai and AI Law Tracker.

Harrison helps teams build AI systems that cut cost and grow revenue. Nearly 20 years of business experience. 2.8M YouTube views. Founder of LaunchReady.ai and the 7 Levels of AI framework. Author of You Have Already Been Replaced by AI.

Connect on LinkedIn