This is a story about a physical therapist named Maria who runs a 3-person practice. It is not a case study with metrics and testimonials. It is a description of what happened when someone with no coding background built a tool for her own business.
Before
Maria's practice runs on a combination of Google Calendar, spreadsheet intake forms, and text messages. Patients book through a scheduling SaaS that costs $120/month. The SaaS does not match how her practice works. It has features she does not need and is missing features she does.
Every morning: 45 minutes reconciling who is coming in, what they need, whether they have specific equipment requirements, whether follow-ups from prior visits need attention. Three tabs open: the scheduling tool, the spreadsheet with patient notes, and a text message thread with her assistant about the day's prep.
The scheduling SaaS's interface is designed for a generic medical practice. Maria's practice specializes in neurological rehabilitation. Her patients often have mobility aids, specific positioning requirements, and accommodation needs that do not fit into a standard scheduling field.
The build
A weekend of Claude Code tutoring sessions from uCreateWithAI. Not a course. Two focused sessions where Maria described what she needed and built it alongside the tutor.
Session 1 (Saturday, 3 hours): The database schema. Patient records with fields Maria defined: name, contact info, diagnosis category, mobility equipment, specific accommodations, and prep notes. An appointment model tied to patients. A daily view.
The CLAUDE.md was configured first. What patient data to never log. What fields are considered PHI and need extra handling. What to never include in error messages.
The intake form came next. Maria described every field. The tutor helped her translate those descriptions into a working web form. By the end of Saturday, patients could fill out an intake form and the data appeared in her dashboard.
Session 2 (Sunday, 3 hours): The appointment calendar view. Not a generic calendar widget. A daily view showing: patient name, appointment time, diagnosis category, equipment needs, and prep notes. Color-coded by appointment type (evaluation, follow-up, discharge).
Then the daily prep checklist: an auto-generated list pulled from the day's appointments showing what equipment to set up, what accommodations to prepare, and what follow-up items are pending from prior visits.
After
The 45-minute morning routine became 10 minutes. Open the app, review the daily prep checklist, confirm everything is ready. The information is in one place instead of three.
The $120/month scheduling SaaS was cancelled. Not because the new tool replaced every feature. It replaced the features Maria actually used.
The important change: when a patient has a specific accommodation that Maria had not anticipated, she added a field for it in 20 minutes during her lunch break. She described the field to Claude Code: "Add a text field called 'positioning requirements' to the patient profile, and show it on the daily prep checklist." It shipped before her next patient arrived.
The tool works exactly the way her practice works, not the other way around. And when her practice changes, the tool changes with it.
What this proves
Maria is not a developer. She does not know JavaScript, React, or SQL. She knows her patients, her practice, and her workflow.
That knowledge is what mattered. The development tool translated her knowledge into working software. The barrier to building was never intelligence or technical aptitude. It was access to a tool that could work with what she knew.
Explore healthcare sessions — our healthcare sessions cover this exact use case.
Get posts like this in your inbox
No spam. New articles on AI strategy, governance, and building with AI for small business.
Keep Reading
5 things health systems consistently get wrong when implementing AI
The five failure patterns that appear in every healthcare AI engagement, and what they cost.
How healthcare COOs should think about AI in their first 90 days
A 90-day framework for healthcare COOs: governance first, training second, tools third.
What a HIPAA-aware patient intake tool actually looks like built with Claude Code
A concrete walkthrough of building a patient intake system with HIPAA guardrails baked into the development process.