Live and in use
Built by me
docandthebone
A practice app for one physiotherapist, built as a gift. From watching her week to a live app on her phone in ten days.
FOR
A physio friend running her own practice around a full-time job
MADE WITH
Claude, Claude Code, Figma, FigJam, Next.js, Supabase, Vercel
TIMELINE
Ten days, late Sept to 2 Oct
ROLE
Solo: strategy, product and design, directing Claude through the build
Why I built it
I built this for a friend of mine, a physiotherapist who runs her own practice, docandthebone, alongside a full-time job. She never asked for an app. On our daily calls I watched her manage her patients from a notes app, WhatsApp threads and memory, and it stressed her out. Booking apps exist, but none of them felt like hers. So I made one that does, as a surprise, and used it to learn the whole journey from an idea to a shipped product.
Who it’s for
One physio, four places, one week
She treats her own patients around a fixed shift: home visits, a clinic slot and sessions in a pool. Every booking is a small sum of time, travel and fees, and her own time is usually the first thing to go.
Build log
It took ten days to go from an idea to her phone
It actually started a week before that. On our calls I kept seeing her work around her own schedule, and that’s where the idea came from.
DefineMapDesignBuild and ship
Decisions
A few calls I had to make along the way
Decision 01
The job sets the shape of the day
Today and the Day view are built around her shift. Sessions sit before and after it, and gaps of 45 minutes or more show as free slots she can book straight into.
Decision 02
Warnings, not walls
If two places are too close in time to travel between, the app says so and offers a better time, but she can still book. She knows her city better than any rule.
Decision 03
WhatsApp, not a messaging system
Her patients already live on WhatsApp. Reminder, running-late and payment messages are templates that open WhatsApp with the text ready. No inbox to check, no business API to pay for.
Decision 04
Her own time is the one thing the app guards
Clashes between sessions are only warnings. But booking over a My time block, like her morning yoga, is blocked unless she overrides it on purpose. The app is relaxed about her schedule and strict about her rest.
Behind the build
App map: every screen, state and edge case, in FigJam before any pixels
Two directions for Today, then the one I chose
The brand board, built out from the logo she already had
Dev notes on every screen: the brief the build ran on
Me and AI
It starts in my head, not in a prompt
By this point I had a way of working. I got her day, the problem and the scope clear in my head first, so every prompt carried the full context. Then I brainstormed with Claude, and nothing it gave me went in without me going through it in depth.
Product and PRD
01
S
Me
Spent a week on calls understanding her day, and got the problem, her and v1 clear in my head before I wrote a single prompt
AI
Claude
Brainstormed it with me and drafted the strategy and PRD from that context
S
Me
Went through every line, questioned the decisions and cut it down to what could ship in ten days
Final callMe
Map and screens
02
S
Me
Worked through her flows and edge cases first, then gave Claude the references, her brand and the rules for every screen
AI
Claude
Drafted the app map, scenario maps and screens in batches, with dev notes
S
Me
Reviewed every batch in depth for missing states, gaps and inconsistencies, and sent it back until it held up
Final callMe
Build
03
S
Me
Broke the build into milestones and wrote a handoff with the full context
AI
Claude
Wrote the code in Claude Code, milestone by milestone
S
Me
Tested every flow on my own phone, caught what broke and made the call to ship
Final callMe
Where it stands
It’s been live on her phone since 2 October
She uses it now and then, not every working day yet, which was the bar I set in the PRD. So v1.1 is about earning that habit: voice notes, a weekly wrap and the Money screen.
What I learned
01
Designing for one person is a sharper brief than designing for a market. Every call had a real answer.
02
Dev notes are design work. Writing down what each tap does caught more gaps than any review.
03
Cutting scope is a design decision. Voice notes and Money waited so v1 could ship on the day I promised.
Want the
full story?
I'm happy to walk you through the files, the trade-offs and what I'd change today. Or tell me about something you're building.
Message me ↗
Based in
Mumbai, working with teams anywhere