I am a Software Engineer working in an Institutional team providing services to professional clients in Forex and Securities.The main focus is on the backend, working in Java Spring and Spring-Boot, and the frontend in React and Typescript on the side.

As a Software Engineer in the Fellowship Origin Programme, I have developed a Web Application for Remote Desktop Service, now actively used by over 4000 CERN users. This application was built using a combination of .NET (C#) for backend operations and REACT with TypeScript for the frontend. For seamless integration, I employed PowerShell scripts for SOAP interactions and automation, deploying the application with Docker and OpenShift. Currently, I am developing the CERN app store, focusing on backend development and optimizing AI models for particle physics research.

01/06/2022 - 01/03/2023 Full-time | 01/03/2023 - 22/02/2024 Part-timeAs a Deep Learning Engineer, I specialize in Computer Vision, Domain Adaptation, and Data Processing using Python, PyTorch, and C++. My focus is on developing infrastructure for autonomous vehicles. This includes working on signal processing and control systems and building an environment for training and testing machine learning models.In my role, I also work on GAN, and Few-Shot Learning networks for semiconductor defect detection. I use Python and PyTorch to develop models for augmenting and classifying semiconductor defects, contributing to quality control and production efficiency advancements. Ultimately, this is a young startup where I also learned how to mentor new members, take on projects, and plan.

Worked on algorithms in C++ and Python for System MCU bootloader.Optimized and integrated DSP algorithms for DSP cores in C/C++.Worked on unit testing in Python and building Jenkins pipelines.

I worked on digital image processing system for measuring angles and angular speed with laser in Python and OpenCV and digital control system with ATML controller for stabilizing pan/tilt system with F.O.G. and accelerometer sensors. The system consists of digital controller created in Matlab and Embedded C downloaded on micontroller and measuring systems created in MATLAB and Python.

I worked on visualizing rocket trajectory with variable acceleration (G forces) and velocities on Google Earth via Python 3. Also created the visual field of radars, which follows a rocket trajectory.

Courses:1) Obect-Oriented Programming C++ held by prof. Dragan Milićev2) Systems and Signals held by prof .Željko Đurović and prof. Predrag Tadić3) Control Systems 1 held by prof. Željko Đurović4) Control Systems 2 held by prof. Aleksandar Rakić5) Nonlinear Control Systems 1 held by prof. Aleksandar RakićJob on this courses consists of assisting students at laboratory, grading their projects/homework and helping Teaching Assistant.

Worked on system that detects and recognizes sounds emitted by quadcopter, implemented on Jetson Nano hardware via python 3 and ROS.First part of the job was creating training and test sets of sound spectrograms.Second part was transforming sounds into spectrograms, then train convolutional neural network with them.Third part was implementing trained network on hardware Jetson Nano and finishing it with ROS.

I designed a controller for a bi-directional buck-boost converter in Simulink, then implemented it on Texas Instruments DSP via C programming language. The work's second part was testing the controller and a converter circuit.