Ruslan Kalinovych

It's my startup with idea to create a radio-controlled car managed by a phone with augmented reality.

I was a product owner and product manager, interviewing people, managing tasks, and communicating with investors and Android engineers who implemented augmented reality with ArCore. Recognized tiny toy cars with OpenCV and wrote core logic using C++.

An OpenGl library for an Instagram story maker app that allows adding a video, GIF, and PNG and placing them freely on a surface.

I implemented a flexible and convenient interface for the client. The library could render added objects into a final video and merge audio streams of video objects on the surface.

A WebRTC and MediaSoup-based video conference app that has particular features for the university.

I started working on the project from the beginning and implemented all WebRTC communication and video stream processing. We had some problems with the MediaSoup library, so we rewrote and recompiled it for our usage. The application had to handle 500 participants with videos and audio; thus, we spent some time optimizing it.

An Android application for football fans created from scratch using clean architecture and Model View Viewmodel (MVVM) achieved a 99.8$ crash-free rate.

Implemented BLE communication with unit tests and in-app Google Billing along with Stripe.

A finance application for creating different types of invoices for businesses. I maintained the project, fixed bugs, implemented new features, improved crash-free users from 66% to 91%, and optimized the app's performance.

A teleprompter application where users film themselves on a frontal camera reading subtitles. The app has multiple features, including recording a video and displaying, adding, editing, and rendering subtitles on a final video with OpenGL.

I created it from scratch following Domain-driven Design (DDD), clean architecture, multi-modules, and MVVM. Used Camera2 and CameraX and implemented a lightweight OpenGl in-app video editor.

Health care applications to track a patient's state of health, communicate with a doctor, and control health devices like blood pressure monitors and oximeters.

The project consisted of four applications that utilized the same core codebase. I maintained the project, fixed bugs, implemented new features, and established communication with the blood pressure monitor and oxygen meter devices.

An app to help teachers check students' answers on test sheets.

Used OpenCV to recognize selected variants and analyze correct answers. My responsibility was to fix OpenCV-related bugs and implement new features.

A cleaner intended for optimizing a device, clearing the system, finding duplicates, and battery saving.

During development, I reverse-engineered other cleaners from Google Play Store to see how they worked and implemented a feature to kill running processes inside the system.

I wanted to create an efficient logging library without impacting performance, even though I could write tons of logs.

Imagine a situation where users could lavishly log the onDraw Canvas method with every single Log.d method call at runtime. We created a StringBuilder object, obtained a string, passed it as an argument to the method, and entered a new stack from the byte code perspective. Thus, we consume CPU time and create a new object, which is not good in the onDraw method. Finally, I came up with an idea for the logging library that makes no objects, doesn't enter a stack, and doesn't have the logging code in the release build.