Erin Downes
61 · veterinarian · PennsylvaniaveTriage — walks receptionists through questions and sorts incoming calls into four urgency tiers.
“Anybody can do this because you just need to try, you just need to be willing to make some mistakes.”
A vet, a farmer, an 89-year-old retiree, a novelist, a new dad. None of them write code. All of them shipped something real with Claude, ChatGPT, Lovable, Replit and the tools you would use in the Lab. Every story here was reported somewhere else first, and every card links to where.
These are not FEZ Media clients or Lab graduates — they are strangers who did it on their own. How this page was put together
Sorted by nothing in particular, because there is no ranking here. A hackathon winner sits next to a dad who built a feeding tracker over two weekends, and both of them count.
veTriage — walks receptionists through questions and sorts incoming calls into four urgency tiers.
“Anybody can do this because you just need to try, you just need to be willing to make some mistakes.”
Plinq — a women’s safety app that checks public records and scores risk green, yellow or red.
“If Lovable didn’t exist, Plinq would never have seen the light of day. I built everything on Lovable — the website, the desktop app, the backend workflows — all without an engineering degree.”
Ireland Cattle Price — pulls live beef prices from factories and marts into one place for farmers.
“I’m only a small farmer but there’s lots of me.”
Crash Out Diary — a place to vent anger safely, with AI personas, breathing exercises and stress-relief games.
“The more AI comes, it doesn’t mean that people need humans less. It just means humans become more of a premium.”
11 free iPhone apps for older adults — including a voice-input helper and a slideshow narrated by his granddaughter.
“I like creating things. When I discovered I could develop apps myself — and if I do, Apple will market them worldwide, just like that — it felt like a great idea.”
Seeing the Difference — shows students the words they chose next to the words an AI would have chosen.
“I—a person unsure which remote controls the television in her own apartment—was able to build a digital tool (one complex enough to require an API key, no less!)”
Tobey’s Tutor — an AI tutor for kids with dyslexia, built for her son, with frustration detection and a parent dashboard.
“I knew what product requirements were, but I don’t know how to code.”
A personal economics dashboard — daily African macro data, company news and digital-economy stats in one screen.
“I never built anything and hadn’t even thought about coding in about a decade.”
Postcard Press — upload a photo, it arrives as a real postcard for $2.
“There is this insane inflection point right now where non-coders can just bring products to life in such a short amount of time with really little learning curve.”
How many layers should I wear today? — a weather app that answers exactly one question: what to put on.
“I vibe coded this with Lovable and use it myself every single day. Weather apps had too much clutter for simple daily clothing decisions.”
A “robot friend” for her daughter’s 4th birthday — bedtime stories, vocabulary help and tooth-brushing pep talks on her phone.
“We’re at the stage where [AI tools] have become very democratized, and you don’t need any technical background.”
Prompt Snips — a site for saving and sharing AI prompts.
“I wouldn’t describe myself as a developer.”
My Baby Logger — tracks feedings, sleep, diapers and meds.
“I’m a new dad, and while waiting for the little one to arrive, I started tinkering around with Lovable. I built My Baby Logger over two weekends.”
CarbScan — snap a photo of a plate, get a carb estimate.
“I built CarbScan to help manage my son’s diabetes and blood glucose levels with faster carb counting. I used Replit. Has become a daily go-to.”
tinyboard — a task manager that shows only three tasks, plus a button that picks a random email to answer.
“software that is so individual that it really only fits one person perfectly. Like a dress made just for you.”
Help Morning — a to-do app with accounts, points and a shop that gamifies her morning routine.
“I want to really understand HTML. I really want to understand JavaScript.”
An AI college-guidance counselor — built in 24 hours at a Cursor hackathon.
“Prompts are supposed to have good details and good information. You have to instruct the AI like a teacher to a student.”
AED-tracking tools — a data-extraction app and a voice web app for a defibrillator company, plus AI classes for seniors.
“Technology isn’t work to me — it’s fun.”
LunchBox Buddy — looks at a photo of the fridge and suggests what to pack for his kid’s lunch.
“I am not a coder. I can’t write a single line of Python, JavaScript or C++.”
No stories used that tool.
Three lines that came up again and again across nineteen very different people. They are also, not coincidentally, three of the habits the Lab is built around.
“I also tell it to tell me one thing at a time, because it can be overwhelming.”
“AI obeys too well, and that has become its own problem. It never says no. I wish it would, sometimes.”
“Try the tools with curiosity and see where it takes you.”
Haris Rana, a 34-year-old physician in California, built a medical-records dashboard with Claude and then could not launch it, because HIPAA compliance is bigger than one person.
“To get it approved would require a lot of strings that you need to pull from very important people. That’s not possible for a random doctor to do.”
Knowing what you are allowed to ship, who it is for, and what “done” looks like is the part that stops most people — long after the code works.
That is why the Lab starts at problem definition, not prompts. Six weeks of it, before anyone opens a tool.
Six weeks, two live sessions a week, six seats. You leave with UX flows, an interface design and a working prototype of your own idea. Designed to be finished.
Every person on this page was written about somewhere else first — in a newspaper, a magazine, a company blog or their own newsletter. Each card links to that source. Every quote is reproduced from it.
Outcomes such as user counts and revenue are as reported by the source, and several of those are self-reported by the builder rather than independently verified. Read them as claims, not audited numbers.
None of these people are FEZ Media clients or Innovation Lab graduates. Some work in tech-adjacent roles — product, marketing, journalism — but none are software engineers by trade. Curated by FEZ Media as evidence that the Lab’s premise holds.