Software Developer in Zagreb, Croatia

Kino.de is the number one German app for finding movies, tv shows, and movie theaters. The app has around 40,000 weekly active users and the crash rate is lower than 0.4%.
While working on this app, I used Objective-C as the main language and I wrote a library for downloading images. The reason I did this was that I wasn't satisfied with the performance of the current assortment of open-source libraries.

For this app, I developed many custom components and one of the most challenging was video player that was inspired by the Facebook iOS app video player (it allows a user to scroll the list of videos and automatically starts video playback when the video scrolls into view).

Crypto Price is a macOS and iOS application for keeping an eye on your favorite crypto tokens. Both applications reuse the same foundations, but they have separate UI files.

During this project, I learned how to reuse code for Mac and iOS and how to properly configure Xcode to work without any problems. This project was written using Swift.

No TV Signal Screensaver is a macOS application that checks user activity and starts a fake screensaver. I say fake because screensaver apps aren't supported by the Mac App Store so an app has to pretend it is a screensaver app. This app is written in Swift and draws the UI using CALayers.

TankenApp is an iOS application for finding the nearest and cheapest gas stations nearby in Germany. App has more than 100,000 weekly active users.

I developed an iOS app and is written 95% by myself. It also required collaboration with Android, back-end, and designer teams.
All code is written in Swift 5, dependencies are managed by CocoaPods and the app is shaped using MVC architecture.
The user interface was created using Xibs and for fast left-right navigation between favorites, the app uses UIPageViewController. Analytics is tracked using Firebase and Amplitude analytics.

The design was provided using Zeplin and all icons inside the app are stored in vector PDFs. VCS used for this project is git and CI used at some point in time was Microsoft AppCenter.

Hasher is a macOS utility application for creating hash strings from the text. It provides support for MD5 and SHA1-512 hash algorithms.
The app is written in Swift using MVC architecture and supports the latest macOS 10.14 dark mode appearance.

Source code of the app is available on GitHub: https://github.com/josipbernat/Hasher

IntelliQuit is iOS application that helps users reduce tobacco usage by monitoring level of tobacco in their urine.
App takes image of NicCheck strip that was in reaction with urine and then makes analysis of given image using computer vision.

My role when creating this app was development and maintenance of iOS app and backend API.
iOS app is written in Swift using MVC architecture, dependencies are managed using CocoaPods and UI is implemented using Storyboards, Xibs and native code.
Camera screen uses OpenCV for detecting required elements inside image. OpenCV is written inside C++ and communication with rest of Swift code is achieved using ObjectiveC++ classes.

Backend APIs are exposed using REST API and backend runs on NodeJS inside AWS EC2 instance. Database used for this project is MySQL database which is hosted inside AWS RDS instance. For communication with other EC2 which does CV analysis I used AWS SQS system. Proxy used for API is Nginx.