QuickNews

After plenty of back and forth with Apple reviewers, we managed to ship the app. We then debugged the multifarious bugs on iOS 13, including notifications. In the meantime we were running Google Ads campaigns for Android. With this, we were ready to introduce it to the wider world.

We completed the first version of the app for Android, and gave it to friends and family to test. Talked to the users constantly, getting instant feedback on new features, quality of recommendations, etc. Used their experiences in guiding our decisions and implementation.

Found a friend of a friend who is experienced with both Android and iOS, and was very motivated to start working on the project. Though still in college, he already had a couple of successful startup projects under his belt, which made him a great fit.

Took Andrew Ng's collection of courses called deeplearning.ai. The most fascinating idea was using vector representations for both images and text data. The text part seemed to be an excellent fit for the news aggregation and recommendation problem which had been in the back of my mind for some time.