Building
Vaulth AI, as its founder and still its only engineer.
The platform runs: seven services on Docker Swarm, 227 endpoints, a mobile app across fifty-one screens in English and five Indian languages, now submitted to the App Store and Google Play for review. Alongside it, a Knowledge Library and Vaulth for Doctors and Hospitals are in active development. My week is consent architecture, AI guardrails, infrastructure cost, and the compliance work that decides what the product is allowed to promise. The case study has the decisions and what each one cost.
This month I’m trying to get from “it runs” to “someone depends on it.” That is mostly a distribution problem, not an engineering one, and I have far less experience with it.
Exploring
Arth is at phase 4 of 5. Phase 5 is blocked on measurement on purpose — its design depends on numbers the app has to collect from real use first, and inventing those numbers to unblock myself would defeat the point of the phase.
Cricket Manager exists and runs. Whether it becomes something other people play is an open question I have not answered.
Separately: how far a deterministic-first cascade can carry a product before a model is needed at all, which is the question underneath both of the above, and one I would like a better answer to than "further than people assume."
Things I’m testing
Three ideas I’m testing in the work, plus one result that keeps me honest.
A small team can still do serious work. Clear product judgment, good tools, and strong checks can let a small team take on more than its size suggests. I’m testing the extreme version by building a healthcare platform with agents and no other engineers, while noting exactly where that stops working.
Solo founders need governance too. Working alone means no code review by default. Vaulth’s engineering rules, blocking CI checks, and claims labels exist because I had to recreate some of the useful friction a team would normally provide.
Some constraints are worth keeping. Arth’s zero-transmission rule and Vaulth’s refusal to claim certification both cost real capability. They also make the products more honest.
The result that makes me cautious. A controlled trial of experienced developers, working in codebases they knew well, found them measurably slower with AI assistance — while predicting beforehand that they would be faster, and still believing afterwards that they had been. That result is inconvenient for all three bets above, which is why the essay argues with it directly rather than around it — and says there what I can and cannot currently source for it. If the first bet is wrong, it will be wrong in that direction.
Next
Getting Vaulth in front of the five people it was designed for — a caregiver managing a parent's records at distance is the one I understand least well and need to be wrong about early. Then the first design partners, then the numbers that decide whether the consent thesis survives contact with people who did not think of it.
I also need to get much better at distribution. It is my weakest skill and probably the one that would change what every future project can become. I want to write more too; one essay is not a thinking practice, and the ideas I’ve changed my mind about are usually the useful ones.
Positions, not predictions — and not a reading list. What I change my mind about lands here when it changes, rather than when it would look good.
If this page ever looks polished, I've stopped updating it.