Skip to content

Sourav Ukil

All work
Live2025

heybarakah.app

A prayer lock app for Muslims

Network calls
None in the critical path
Conventions
Four juristic methods
Status
Live

The problem

Most prayer apps are trackers. They are very good at telling you how much you missed, which is a report on failure delivered daily. The gap was an app that helped someone actually pray rather than one that helped them measure whether they had.

What I built

A lock screen that stands between you and the phone at prayer time and does not move until the window passes or you mark the prayer done. Consistency was the whole product goal, so every screen was judged on one question: does this help someone pray, or does it only help them feel informed?

An app that shows a spinner at maghrib is not a slow app. It is a broken one.

The interesting part

Prayer times are computed locally, not fetched, because the app has to be right in a tunnel and on a plane. That meant implementing the astronomical calculations and the four main juristic conventions rather than leaning on an API, and it removed a network dependency from the one moment the app cannot fail.

The edge case that took longest

Above roughly 48 degrees latitude there are summer nights where the sun never drops far enough for isha to have a real time. There is no correct answer for what to show, only conventions different communities follow, so the app asks once and then never asks again.

Outcome

Live. The metric that matters is not installs, it is whether someone still opens it in month three. That is the only number I look at.

What I would do differently

I would cut the statistics screen entirely. It was the first thing I built and it is the thing that pulls the product back toward being a tracker, which is the category I was trying to leave.