Matthew Gallagher will tell you he spent $20,000 and two months building a telehealth company. What he won't tell you—at least not right away—is that by the spring of 2026, MEDVi was tracking toward $1.8 billion in annual revenue with exactly two full-time employees: himself and his brother.
The math seems to defy basic arithmetic. Until you realize Gallagher wasn't really building a healthcare company at all. He was building a marketing layer on top of someone else's healthcare infrastructure, automated to within an inch of its life.
The 41-year-old Los Angeles entrepreneur launched MEDVi around September 2024 with a bet that felt almost too obvious: the surging demand for GLP-1 weight-loss drugs had blown open a gap in the market, and if you were willing to strip healthcare delivery down to its digital skeleton, you could slip right through. By the time The New York Times reviewed the company's financials in early 2026, MEDVi had served roughly 250,000 customers and generated $401 million in revenue during its first full year—most of it profit.
It reads like the extreme edge of what becomes possible when AI automation collides with outsourced clinical infrastructure. It also reads, depending on your vantage point, like exactly the sort of thing regulators lose sleep over.
When AI Isn't an Assistant—It's the Team
Gallagher came to MEDVi from the watch subscription business, of all places. Watch Gang, which he'd previously founded, taught him funnel optimization and customer acquisition. What it didn't teach him was how to code, design ads, or handle customer service at scale. For MEDVi, he didn't hire people to fill those gaps. He auditioned AI tools.
ChatGPT and Claude handled the coding. Midjourney and Runway generated ad creative. ElevenLabs produced voice work for communications. Intake forms, the customer support chatbot, operational scaffolding—all of it built through what one tech newsletter described, perhaps generously, as "vibe coding." Iterate with the AI until it works. Ship it. Repeat.
This wasn't a case of a founder using AI to move faster with a lean team. This was AI as the team, with Gallagher conducting an orchestra of more than a dozen tools to handle functions that would typically require dedicated staff. Marketing campaigns materialized from prompts. Customer inquiries got fielded by bots that had learned to sound vaguely human.
The infrastructure stayed skeletal because it had to. MEDVi launched without venture funding, bootstrapped entirely from that initial five-figure outlay. Which meant every dollar that didn't go toward payroll could go toward Facebook ads.
The model only worked—and this is the crucial part—because Gallagher outsourced the expensive, heavily regulated pieces of healthcare to companies already running that machinery. Clinical consultations, prescription processing, pharmacy fulfillment, compliance oversight: all of it ran through partners like OpenLoop Health and platforms such as CareValidate's CareGLP. White-label telehealth infrastructure that let MEDVi operate as a consumer-facing brand without employing doctors, building pharmacies, or navigating the byzantine world of healthcare regulation directly.
In effect, MEDVi was a storefront. The actual healthcare happened elsewhere.
The Numbers That Shouldn't Work (But Did)
In 2025, according to financials the company shared, MEDVi pulled in $401 million in revenue with a 16.2% net margin—roughly $65 million in profit. By April of the following year, daily sales were running north of $3 million.
To put that in context: venture-backed digital health companies spend years and hundreds of millions in funding trying to reach that kind of scale. Ro, which went public via SPAC, took more than five years to hit comparable revenue numbers. Hims & Hers, now publicly traded, raised over $200 million before achieving similar traction.
MEDVi did it in eighteen months. With two people.
The product itself was straightforward, almost aggressively simple. Compounded semaglutide and tirzepatide—the same active ingredients in Ozempic and Mounjaro—delivered through a telehealth funnel that started at $179 for the first month, $299 for refills. By March 2026, MEDVi operated in 49 states, riding a wave of consumer demand for weight-loss medications that had been simultaneously in shortage and financially out of reach through traditional channels.
The economics worked because the overhead barely existed. No medical staff on payroll. No brick-and-mortar facilities. No sprawling call center. The company's website even disclosed, in language that suggested either transparency or legal caution, that "certain materials on this website may be generated or enhanced using AI technologies."
Whether customers read that disclosure is another question entirely.
The Machinery Behind the Brand
MEDVi's fine print tells a more complicated story than its headlines suggest. The site named OpenLoop Health as the clinical network powering provider consultations. Prescriptions flowed through entities identified as "CareGLP Affiliated P.C.s." Other pages listed Beluga Health and Belmar Pharmacy as partners—implying multiple infrastructure stacks running in parallel, possibly to manage volume or regulatory fragmentation across state lines.
None of this is particularly unusual in the world of GLP-1 telehealth. The boom in weight-loss drugs has spawned an entire ecosystem of white-label platforms designed to let consumer brands launch quickly without building clinical operations from scratch. CareValidate and OpenLoop essentially offer plug-and-play infrastructure for companies that want to sell compounded medications without the burden of employing prescribers or owning pharmacies.
MEDVi, in essence, franchised the regulated parts and built a marketing engine on top.
The strategy let Gallagher focus on what he actually knew: customer acquisition, brand building, funnel optimization. By February 2026, MEDVi had expanded beyond weight loss into men's health with a compounded erectile dysfunction product it called QUAD. Meal delivery was in the works. Women's health, hair loss, and skin treatments sat in "coming soon" status on the homepage, placeholders for future revenue streams.
It looked, from a distance, like a company preparing to become a full-fledged health and wellness platform. Up close, it looked more like a series of landing pages connected to the same backend infrastructure.
When the FDA Comes Calling
On February 20, 2026, the Food and Drug Administration sent MEDVi a warning letter. The agency flagged what it described as misbranding in the company's marketing and labeling of compounded GLP-1 products, requesting corrective action within 15 business days. The letter arrived six weeks before the Times profile would introduce MEDVi to a broader audience beyond the startup-watching corners of the internet.
It also signaled that the company's rapid ascent had attracted attention from precisely the people Gallagher had been trying to avoid.
The FDA warning wasn't MEDVi's only brush with legal scrutiny. A consumer class action filed in November 2025 named the company among several GLP-1 telehealth brands operating through the OpenLoop ecosystem, alleging deceptive marketing practices around so-called "oral tirzepatide." The case was filed but hadn't been adjudicated. Earlier, in June 2025, a media outlet had criticized MEDVi for allegedly using AI-generated before-and-after photos and publisher logos without actual linked coverage—accusations that raised questions about how aggressively the company was deploying AI in customer-facing materials, and whether anyone was checking the output before it went live.
Consumer complaints on platforms like the Better Business Bureau and Reddit painted the kind of mixed picture you'd expect from a company growing this fast with this little human oversight. Some customers reported positive experiences: fast shipping, effective weight loss, responsive (if automated) support. Others complained about delivery delays, refund disputes, and the distinct sense that no one was actually reading their messages.
That's the thing about replacing human judgment with automation. It works until it doesn't.
What MEDVi Actually Proves
MEDVi is an outlier. Maybe too much of one to serve as a blueprint for anything beyond MEDVi itself.
The GLP-1 shortage created a narrow, temporary window where demand vastly outstripped supply and regulatory gray areas allowed compounding pharmacies to step into the breach. Gallagher threaded that needle with remarkable timing, building a business that could scale as fast as ads could run. The company's trajectory through 2026 suggests he caught lightning in a bottle.
The FDA warning letter suggests that lightning doesn't strike twice in the same place—at least not without consequences.
The broader question MEDVi raises isn't whether two people can run a billion-dollar company. It's what AI-enabled business models look like in regulated industries where the barriers to entry were historically defined by capital and compliance. Gallagher demonstrated that it's now possible to generate significant revenue with almost no human infrastructure. But he also demonstrated the limits. You can automate marketing and operations. You can't automate away the FDA. You can build a brand without employees, but you can't ignore the fact that you're selling controlled pharmaceuticals, not watches.
For founders watching MEDVi's arc—and there are many—the lesson isn't about replicating the model. It's that the pieces now exist to test business ideas at speeds that were impossible even five years ago. Launch in weeks instead of years. Scale to tens of millions in revenue before raising a dollar.
And discover, perhaps faster than you'd like, that success at that velocity brings regulatory scrutiny that moves just as quickly. Gallagher built MEDVi in two months. The FDA found it in seventeen.
