Why I started building ITISYOU

I did not set out to build a network. I set out to build one calculator, because I could not find one that worked the way I needed it to, and everything after that happened by accident and then by decision.

I studied computer science in Tirupati, finishing my Bachelor of Technology in 2024, and moved to the UK that September for a masters in Artificial Intelligence at Liverpool John Moores University. I finished it in September 2025. What I took from the degree was not really the models. It was the habit of asking what a result actually shows, which turns out to be a more transferable skill than any particular architecture.

By June 2026 I was in Chester and building in the hours around shift work. That is still true today, and I mention it because the shape of this project comes directly from it. When your building time arrives in two-hour blocks after a shift, you learn very quickly which parts of software development are the actual work and which parts you had assumed were the actual work.

The first thing I built was not the hard part

The first tool was a shift pay calculator. I needed to know what a particular week would pay once unpaid breaks, night premiums and weekend premiums were accounted for, and the answer was genuinely non-obvious. The calculation itself took an evening. It was correct, it was fast, and it was completely useless to anyone but me, because it lived on my laptop.

Everything after that evening was the real project. Where does it live. What happens when someone enters something strange. What does it do on a phone with a bad connection. What happens if it is wrong and someone plans around it. Who is responsible for that. Those questions have almost nothing to do with the calculation and almost everything to do with whether a thing is a product or a script.

The calculator became a collection of fifteen of them, because once the surrounding machinery exists, the marginal cost of the next tool is small and the marginal usefulness is not. That was the first accident.

The list of things I did not know

I had expected to spend the year on engineering problems. Instead I spent a large part of it on a list of things that were not engineering problems at all:

• Design. Not decoration — hierarchy. Deciding what a person should read first and making everything else quieter is harder than it looks and cannot be delegated to a component library.
• Privacy and data protection. I assumed that collecting almost nothing would keep me out of the obligations. It does not. That turned into its own lesson.
• Accessibility. Keyboard operation, focus, contrast, motion. Retrofitting these is miserable; building with them is barely any extra work. I learned that in the wrong order.
• Cost. Every product here has to run on a budget that one person’s wage can absorb. That constraint has improved the architecture more than any principle I chose on purpose.
• Quality judgement. Knowing when something correct is still not good enough. This is the one I have found hardest, and the one that has cost me the most rebuilt work.

Two products, then five

The second product came from the opposite instinct. Space is a quiet screen with nothing to achieve on it. I built it because I had spent months inside tools designed to keep me engaged and wanted one that was designed to let me leave. It has no streak, no leaderboard and no message about the day you missed, and every one of those absences was a decision I had to make deliberately, because the defaults all point the other way.

Space now charges: £1 for a month, or £10 for a year. They are one-off payments rather than renewing subscriptions, and the renewal was taken out deliberately — an automatic charge someone forgot about is the most expensive kind of customer to have, in fees and in trust. It means asking people to come back and pay again, which is harder. I would rather that than the alternative.

The other three came from problems I hit while building the first two. DEV.ITISYOU exists because I lost an afternoon to a highly upvoted answer that had been correct on a version I was not running. ATSYou exists because I was applying for engineering roles at the same time as building, and every CV tool I tried was willing to improve my history in ways I had not asked for. ITISYOU News exists because I kept losing the thread of stories I had stopped following, and no reader I used could tell me what had changed since I last looked.

That is five, which is more than one person should probably maintain. I am aware of the objection. My answer for now is that they share almost all of their infrastructure and about half of their ideas, and that the alternative — one product, polished for four years, launched into a market I had not tested — did not obviously look wiser.

What I have refused to do

The most useful thing I have learned is how to throw away work that functions. Space had a cinematic entry sequence. It was built, it ran, and it was not good enough, so it is gone — and it is not described anywhere on this site as a feature, because it is not one.

ATSYou is the larger version of the same decision. The generator works. It produces documents where every line traces to something the user confirmed. It is still not open to customers, because the editorial and layout quality is not where it needs to be to charge someone for it. I wrote separately about why that was such a hard call.

Where this actually stands

Three of the five products are usable today. One is publicly reachable but has not opened to customers. One is built and running but has not launched. The status page lists all of that with the dates each was last checked, and it will say so even when it is unflattering.

I am still building, and still applying for full-time engineering roles in the UK. ITISYOU is not a success story I am reporting after the fact. It is the work in progress, published while it is still in progress, which is a slightly uncomfortable way to do it and the only version I can write honestly.