😸 AI did WHAT for $20K?

2026-04-06 · Source: The Neuron · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Entrepreneurship & Start-ups, Emerging Technologies & Innovation · Depth: Novice, long

Summary

Matthew Gallagher launched Medvi, a telehealth startup, in September 2024 with a $20,000 investment, a dozen AI tools, and initially zero employees. Within 18 months, Medvi is projected to achieve $1.8 billion in sales, primarily leveraging AI for coding, website content, ad generation, customer service, and business analytics. Gallagher utilized tools like ChatGPT, Claude, Grok, Midjourney, Runway, and ElevenLabs, alongside "telehealth-in-a-box" platforms CareValidate and OpenLoop Health, to streamline operations. Despite initial "AI slop" in marketing and some chatbot errors, the company generated $401 million in revenue in 2025 with a 2% net profit margin and served 250,000 customers by year-end. The article also notes Microsoft's Copilot terms of use stating it is "for entertainment purposes only" and details how Google's Gemma 4 model can run locally on a MacBook at 51 words per second.

Key takeaway

For CTOs and VPs of Engineering evaluating new ventures or operational efficiencies, this case demonstrates that a lean, AI-first approach can yield substantial revenue quickly. You should consider integrating a diverse suite of AI tools for rapid prototyping and core business functions, understanding that initial "AI slop" is acceptable if it enables market entry and revenue generation that can fund subsequent refinement and human-led upgrades.

Key insights

AI tools enable rapid, lean startup creation and scaling, even with minimal initial investment and human capital.

Principles

• Good enough initial solutions can fund later upgrades.
• AI fluency combined with grit drives significant business outcomes.

Method

Utilize "telehealth-in-a-box" platforms for core services, then deploy diverse AI tools for front-end, marketing, customer service, and operational automation.

In practice

• Integrate AI agents for automated system communication.
• Run large language models like Gemma 4 locally via LM Studio.
• Use AI for initial content generation, then refine with human oversight.

Topics

• Microsoft Copilot Disclaimer
• AI-Powered Startups
• Telehealth Innovation
• Local LLM Deployment
• Gemma 4

Code references

• JuliusBrussee/caveman
• salmanmohammadi/nanocode

Best for: CTO, VP of Engineering/Data, Director of AI/ML, Entrepreneur, AI Student, General Interest

Related on AIssential

• TAI #194: AI Goes Macro; Job Loss Fears, Military Usage, OpenAI $110B Raise — AI's macro impact is accelerating, driving unprecedented investment, geopolitical tensions, and job market shifts.
• Exclusive: Index Ventures leads $20m round in AI inspection startup Scope — AI-powered computer vision enhances industrial visual inspections, reducing human error and boosting efficiency in a $300bn market.
• Sources: Cursor in talks to raise $2B+ at $50B valuation as enterprise growth surges — AI coding startup Cursor is securing significant funding and revenue growth despite competition, driven by strategic model diversification.
• AI just made the billion-dollar solo founder real — AI tools enable rapid, lean startup scaling, making solo billion-dollar companies a reality.
• Claude and Codex Go to War — Alphabet's Q4 2025 results highlight strong AI-driven growth across its core businesses and significant future investment.
• Intel Delivers Strong AI-Fueled Outlook — AI-driven demand is reshaping tech giants' strategies, leading to both unprecedented investments and significant workforce adjustments.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by The Neuron.