About
It didn't start with a business plan or a pitch deck. It started with a tax return.
When I was doing my taxes earlier this year I looked back and realized I'd gotten my ChatGPT account in August of 2024. For the first six months or so I used it like most people do — a glorified chatbot that talked back to me. Honestly impressive, but I hadn't figured out what it was really for yet.
Then something shifted. I started realizing it could solve actual business problems I was staring at every day. That's when everything changed.
By July 2025 I was in a car with my cousin showing him it could hold a full conversation — then showing him it could write code, connect programs, build automations. Something clicked for both of us. He had an IGBT parts business selling components all over the world. He had data in silos — QuickBooks, Shopify, warehouse inventory in Google Sheets — all disconnected, all manual. And he had $1,000 a month for someone like me if I wanted to help.
So we started building on weekends.
That was the beginning of AMG AI Solutions. Not a grand vision. A car ride, a curious conversation, and a cousin who needed help.
I've been working since I was 13. Started on Captain Anderson's boat in Clearwater, became a first mate at 15 on the Caladesi Island Connection — the first person to ever hold that role. Built pools. Did plumbing. Got promoted from plumber to Plant Manager at United Modular overseeing 235 employees before I was in my mid-20s. Spent years in electrical contracting, construction, roofing, HVAC, and solar.
I've never been the developer or the engineer. I've always been the person who sat in the real rooms with real business owners and figured out how to make things actually work.
In September 2021 I joined a multi-brand home services company — Solar Max, Ridge Max Roofing, Velocity Air Conditioning — and spent 4.5 years building their entire digital and operational backbone from scratch. CRM systems built from zero. 19,000+ automated tasks running every month. A $5M revenue pipeline created from one automation that connected two companies' data. AI quality assurance tools for calls and job photos. A predictive dialer campaign generating $1M+ a year off one $150/month seat license.
I did all of it — and when something needed to be connected, moved, or wired up, that was me too. Workstations, phones, network connections — whatever it took to keep operations running. I had someone handling the internal hardware on the PCs, but everything else that needed to be plugged in, set up, or figured out? That landed on my desk.
As I got deeper into automation, I watched my own skills evolve in real time.
First it was Zapier — no-code automation, drag and drop, 19,000 tasks a month. Then I realized I could cut out the middleware and talk directly to APIs using Google Apps Script. Faster, cheaper, more powerful. Then AI came along and changed the game entirely — and I realized the integration layer was about to disappear. MCP connectors let the LLM understand the data structure itself and query it natively. Point it at a source, describe what you want, it figures out the rest.
Three generations of integration thinking, lived in sequence.
I started going to AI events in Tampa. Found a podcast I respected — Nate B Jones, real utility AI content, no hype. Got a Chamber of Commerce membership for AMG AI Solutions and started showing up in the community. Started trying to help local businesses figure out what AI could actually do for them.
On April 21st, 2026 — one week after getting laid off — I sent an email blast to 5,500 Chamber of Commerce contacts asking one honest question: what does AI actually look like inside your company right now?
One person replied. Someone I already had a professional relationship with. It was good to know they were paying attention — but that was it.
Nobody wanted to talk about their dirty data. Nobody wanted to admit the shadow AI happening all around them. And honestly I get it — the moment you acknowledge it, it starts costing money to fix it. It's easier to keep accepting the pain until the pain forces the change.
What I took from it wasn't discouragement. It was clarity. The only way to actually help companies with this stuff is to be inside them — where the trust already exists, the problems are visible, and someone can actually do something about it.
On April 13, 2026, I got laid off. The solar industry had taken a gut punch from policy changes and the company was bleeding. It stung — but honestly, I'd been ready to move on for a while. I'd grown the business significantly, built things that made real money, and never felt like any of it was truly valued.
That same evening — seven hours later — I got an email from NVIDIA telling me I'd been accepted into their Connect program.
I took that as a sign.
Three days later I was building.
I started Friday night, April 17th. By Monday morning, April 20th — 26 hours later — I had a full private AI gateway running on a GH200 superchip. The kind of infrastructure companies pay enterprise money for. NVIDIA NIM running Llama 3.3 70B. AnythingLLM frontend. Docker containers. Cloudflare Zero Trust. Grafana and Prometheus monitoring. A working product built to understand what enterprise AI actually feels like from the inside.
I built it not to sell it but to understand it. It changed how I think about everything.
→ Read the full build storyI spent a couple of years trying to help small and medium businesses build AI workflows from the outside. The honest truth? They're moving slow and I can't afford to starve waiting for them to catch up.
So I've changed how this site looks. Changed how I'm showing up. Changed my position and my approach.
I'm not trying to be a guru or build an agency. I'm a curious person with 25 years of operational experience, a hands-on NVIDIA build under my belt, active consulting clients, and a genuine belief that the companies willing to embrace AI right now are going to separate from the ones that aren't.
I want to be inside one of those companies.
AMG AI Solutions stays — small LLC, NVIDIA Connect membership, a podcast coming, and a couple of consulting clients I'm still actively helping. But it's a side project and a learning platform, not a competing business.
The NVIDIA Connect program believed in me. I intend to keep earning that. I'll keep building with their model credits in my own time, keep learning, keep sharing what I find.
This isn't the end of anything. It's just a change in season.
If you're not afraid of AI, you want real growth, and you're looking for someone who'll ride hard for your team — get on the magic carpet ride.
Matt Gamble · Clearwater, FL
If you're not afraid of AI, you want real growth, and you're looking for someone who'll ride hard for your team — get on the magic carpet ride.
Let's talk →