The thesis

The future of veterinary medicine is already running in a clinic in Houston.

I'm a solo vet in Houston. I started by asking an AI to fix a Facebook post. Three years later I'm running an AI scribe in my exam rooms and building the tools myself, with Claude and Cursor doing most of the typing. I didn't plan any of this. I just kept hitting problems AI could solve, and the distance from idea to working software kept shrinking.

Dr. Steve Pelton in an exam room at Hearthstone Animal Clinic, with Copper the puppy on the table FP.VET · HOUSTON
Dr. Steve Pelton, DVM · Houston, TX
Where this is going

the same twelve questions

If you've got a clinic website that's basically a page, your hours, and a phone number, that was the right call for a long time. Mine was the same. It worked because people searched, found you, and read it.

That's the part that's changing. More and more, people don't read your site at all. They ask an assistant, and it answers for them using whatever it can find about you. It's answering for your clinic right now, whether you helped it or not.

The phone is next. Most calls to a vet clinic are the same twelve questions, and a machine that never has a rough morning can handle a good number of them. I'm not predicting that. An AI already fields calls at my clinic, today.

None of this replaces the medicine, and it doesn't replace judgment. It replaces the typing, the phone tag, and the after-hours voicemail nobody calls back. The vet who learns these tools doesn't get replaced by AI. The vet who skips them just ends up competing with the one who didn't. That's the whole thesis. The rest of this site is how.

The part people miss

If you've only ever asked ChatGPT, you've seen about five percent of what this can do.

That's not a knock. A chatbot is where most of us started, me included. But a chatbot answers questions. The tools I actually use now, Claude and Cursor mostly, with Lovable when I want a screen fast, sit at my computer and do the work. Lately I run them together, and that's when it stopped feeling like a trick.

Claude Code has become my IT department. I can ask it to check the CPU temperature on the clinic server, and it does. It built the monitor that watches my network. It walks me through a printer that stopped talking to the front desk. Somewhere past 99 percent of the tech problems in my clinic this year got solved by describing the problem and letting it work.

There's nothing special about me or any of this. The whole method fits in one line. What's the issue? Ask Claude. I don't know how to approach this? Ask Claude. If your clinic has a problem, or a task you repeat every day, you can probably build something that fixes it or makes it easier. You don't need to be technical to start. You need to take the time to learn, and most people won't. That's the gap.

"Check the CPU temperature on the server."
It logged in and read it.
"Something on the network keeps dropping."
It built a monitor. It's still running.
"The front desk printer stopped talking."
It found what changed and walked me through the fix.
"Write the discharge the way I'd say it."
That one turned into a product.

Or just ask me

Stuck on something? Ask.

No account, no forum, no form letter. Tell me what you're trying to do or what's broken and I'll tell you how I'd go at it. If it's a good one it usually turns into a thread in the room, or a tool.

Goes straight to me, not a support queue. I answer these myself.

Things I actually built and use

Not a demo reel. Tools I run in a working practice, plus a couple I built for fun.

All 67, including the ones that died →
Running daily

Client comms in my voice

Owner explanations, discharge notes, and handouts written from a line of my shorthand, not from a template. It's the thing I use most.

Running daily

An AI receptionist on the line

Call flows I wrote for emergencies, refills, and new clients. The real work was deciding what it should handle and when it hands off to a human.

Running daily

A scribe in my exam rooms

I talk through the exam. It drafts the note and the discharge and puts them back in the record. Running on my own hardware.

Experiment

Practice data past the screens

Experiments pulling real clinic data through my practice software's API, instead of pasting text in. Still figuring out where it goes.

Live

DadCaptions

A live-caption app I built in a weekend for my dad, who loses the thread in a loud cabin. Not veterinary, but it's proof the wall came down.

Died

Rocky, a voice assistant

Ran on the surgery PC until it didn't. Version one of most things dies quietly. Its replacement lives on its own machine now.

Some of these are solid daily tools. Some are experiments I'm still poking at, and I'll always tell you which is which. Two of them went home with real patients today, and you can read the actual discharge sheets. One of them you can go use right now, no login.

The honest part

Things AI still sucks at.

The demos are amazing. Tuesday afternoon is harder. If a site only shows you the wins, it's selling you something. I've worked around most of these. It just takes some tweaking, and none of it fixed itself.

  • Remembering everything correctly

    Fixed A memory layer it can read and write. Took longer than any of the tools did.

  • Knowing when its own answer is subtly wrong

    Still open I'm the filter. I don't expect that to change, and I'm fine with it.

  • Understanding a workflow constraint you never said out loud

    Fixed I said it out loud. The clinic's real context lives in a file it reads every time.

  • Making a printable that's actually readable on the first try

    Fixed Templates. The first try is still ugly. The second is fine.

  • Staying out of generic marketing voice

    Fixed A voice profile. It reads my own writing before it writes anything.

  • Producing code that works every single time

    Still open It won't, ever. Tests and a rollback are the fix, same as for people.

What this is

Not a newsletter. A room full of vets figuring this out, with me in it.

Tell me what's broken

Sign up and describe the problem in your clinic, or the task you're sick of repeating. I read every one. If it's interesting and enough people share it, I may build it with you, or for you, because I enjoy this more than is reasonable.

The room is open

A private room for vets, techs and managers. Reading is open to anyone. Posting takes a free account. Ask how I did something and get a real answer instead of a course pitch. Free while it's small, and I'll say so before that ever changes.

Watch it get built

I'll post a video of one of my tools. You'll say "I want that for my clinic." Then we talk about how: Lovable, Claude, Cursor, or describe it and I'll see what I can come up with.

One more idea I'm turning over: a version of me you can ask questions, that knows everything on this site and how it was built. Not built yet. If enough of you want it, it will be.