Ten years of building things that had to work.

In 2015 we started PRS India for the unremarkable reason that we thought we could do the work better than the people we had been doing it for. There was no thesis and no funding. There was a desk in Sun Light Colony, two engineers, and thirty projects in the first year — most of them brochure sites for Delhi businesses who wanted to be online by Diwali.

The turn came in 2017. A freight operator in West Delhi asked us to replace the whiteboard their dispatchers used to assign trucks. It was not a website. People would sit in front of it for nine hours a day and their working lives would be better or worse depending on how we built it. We got a lot of it wrong the first time, rebuilt half of it, and shipped something that is still running today.

That project taught us the thing that has shaped every year since: there is an enormous difference between software that demos well and software that survives contact with the people who have to use it. Most of the industry sells the first and quietly hopes you will not notice the difference until the invoice has cleared.

So we built the studio around the second. A principal engineer reviews every architecture before a line of production code is written. Every system is load-tested before launch, not after. Every handover includes an architecture document, tested backups and a session with your own engineers, because a project we cannot leave is a project we did not finish.

It makes us more expensive than the shop down the road. It also means about a third of our work now comes from teams who hired that shop first.