Salt · A family story · 2026 – ongoing
Running the Dad test
Billing and inventory software for my dad’s medicine shop. It ships only when Dad approves.
Salt is billing and inventory software for medicine shops like my dad’s: Indian medical retail and wholesale, chemists, surgical, ayurveda. I am building it myself, first for his counter. The demo is live with dummy data; log in with Gmail and run a billing counter yourself.
Origin
The plan was always Dad’s shop
I grew up around my dad’s wholesale medicine counter, and working beside him someday was always the plan. So when I left my job, with five years of building software for big companies and an executive MBA under way, the question was never what to build. It was whether I could build it alone this time. Salt is the first thing I made after leaving: everything the job and the degree taught me, put to work on the business I know best.
The bar
Dad is my harshest tester
My dad has run his wholesale medicine shop for decades, and he treats software as a necessary evil. So the Dad test is simple: Salt passes when he runs the shop on it, not when he says it is nice.
He has earned that caution. His software dates from the DOS days and lives on one fragile laptop. When the shop gets stuck, his holiday gets a phone call. And every single evening ends with the shop’s whole life copied out to a pen drive. My answer is to automate the heavy work until the software stops asking for his attention, and to give him back his evenings.
That ritual continues today. That is the bar.
The moment
It found dues Dad had forgotten
The first real test was moving 8 years of the shop’s history into Salt: thousands of ledgers and bills. A tool I wrote for the job read them out of the old software’s files. It came through at zero loss, on the one set of records I could not afford to lose. Then Salt started reporting things back, including people who still owed money that even Dad had forgotten.
He finally understood what I do, and why I do it.
Decision 1 · The counter
Built for the rush
Nothing from my years of office software looked like a medicine shop’s billing counter, so I learned the design by standing at ours. The staff’s fingers already know the old keyboard shortcuts, and Salt keeps the same keys. Watching also gave me the real requirement: roughly 250 orders a day across three counters, phones ringing, customers waiting. Software for offices assumes a desk and a quiet hour. The counter has neither.
A busy shop cannot slow down for its software.
Decision 2 · What Dad said
Good was not enough
My dad has run everything in Salt that ships today. His verdict: good, and still not a reason to switch. His current tool is not just software. It is years of habit and small workarounds that already fit his hands. Any new product, however well designed, asks him to relearn his day before it gives anything back. I design software for a living, and the most useful thing this project has taught me is the limit of my own craft.
Design alone is not enough to make them switch.
Decision 3 · The way in
A helper first, the counter later
So Salt does not ask my dad to drop his habits on day one. It sits beside his current tool and does the jobs that tool never did: filing the taxes, suggesting how much of each medicine to order next. Useful first, the billing counter later. Next on the list: purchase bills that read themselves in, and stock that works out its own reorders. Then a voice assistant that takes phone orders and holds each one to a promised delivery time, with no person on the line.
What’s inside
The whole shop in one system
The billing counter is keyboard first and also works by touch. Stock is tracked batch by batch: Salt knows every expiry date, sells the older stock first, and raises its own purchase orders when something runs low. The rules of the trade are enforced by the system, not by memory. Schedule drugs, GST invoice numbering and the regular filings are all handled on the server, where nobody can skip them. A family running more than one shop sees them all in one report. And everything stays as simple as the apps on a phone, because I want the software to be the easiest part of the shop’s day.
Everything Salt does today took five days to build. Design happened in code the whole way; I never opened Figma. The loop is simple. Sit with the people who run the shop, give that understanding to Claude Code, and use the result myself. Then put it in front of real hands at the counter, and listen.
A live reconstruction with dummy data. Bill a customer and watch Inventory and Parties update; every bill is printable.
Today · Honestly
The test is still running
No active users yet; the Dad test is still in progress. The first win was one I never demoed: my dad now carries his work home, moving between laptop and phone, without the old anxiety. It happened on its own, which is exactly what I was hoping the design would do.
The pen drive is still part of every evening; that ritual has not stopped yet. When Salt finally replaces it, the work will not stop at one shop. Every small shop on the street has the same counter and warehouse needs, and that is the change I am building toward.