CV

Fully-Funded PhD Researcher via the Knowledge Foundation (KK-Stiftelsen)

Mälardalens University (IDT - School of Innovation, Design and Technology)

Computer Science. Ongoing project: PerFlex: Performant and Flexible digital Systems through Verifiable AI.

• Architected and implemented an automated vehicular scenario generation pipeline utilizing the esmini simulation environment. Integrated a novel HASCo (Hybrid AI Scene Compiler) framework to orchestrate complex traffic scenarios for the validation of autonomous drive features under industrial safety constraints.
• Engineered a Deep Learning-based cache simulator using LSTM networks to predict cache miss distributions, achieving near-native accuracy while significantly optimizing execution time compared to tools like DynamoRIO.
• Developed a CPU load forecasting framework for industrial embedded systems, implementing Conformal Prediction for uncertainty quantification and Shapley values for real-time interpretability on resource-constrained hardware.
• Architected a full-stack Electron/Typescript application to streamline AI model deployment, integrating Docker containerization for reproducible testing across simulation environments.
• Co-authored a neural network reduction and verification algorithm building on top of the formal verification tool Marabou.
• Collaborating with industrial partners (Volvo Cars, Hitachi Energy, Ericsson) to align research with production-grade safety standards, and supervised multiple Master Thesis students in AI/Reinforcement Learning topics.

Software Engineer, Control Systems Laboratory

Cosylab

Engineered and took responsibility for several software components of a high-level radiation therapy control system.

• Engineered critical Java modules for radiation therapy control, architecting a unit safety analysis framework that ensured regulatory compliance for clinical facilities.
• On-site engineering for an international client medical facility in Helsinki, Finland.
• Successfully coordinated with international clients and technical experts.

Student Developer, Control Systems Laboratory

Cosylab

• Contributed to the development and software engineering of commercial projects using C#.
• Assisted in project architecture implementation of SCADA protocols within a C# architecture.

Master's Degree

University of Ljubljana, Faculty of Electrical Engineering

Department of Robotics

• Master Thesis: Dynamic Movement Primitives in High-Precision Robotic Applications (Presented at the 30th International Electrotechnical and Computer Science Conference ERK 2021).
• Analyzed dynamic movement primitives as an algorithmic trajectory generation and reproduction tool.
• Developed a new second-order filter approach to improve the precision of dynamic movement primitives on the example of camera-guided bacterial colony extraction.
• Utilized MATLAB and Simulink, along with C++ PD controller.
• Robotics control, forward and inverse kinematics and dynamics, signal and sensor theory studied throughout various courses and practical implementations.
• Collaboration with ABB Robotics in Ljubljana, with ABB RobotStudio utilized on a novel pick-and-place implementation based on multiple laser distance measurements to target.

Bachelor's Degree

Sarajevo University, Faculty of Electrical Engineering

Department of Automatic Control and Electronics

• Bachelor Thesis: Lazy Probability Roadmap Path Planning Algorithm for Robotic Manipulators.
• Analyzed, described and implemented a graph-based Lazy PRM algorithm for path planning.
• Implemented the algorithm in C++ (Boost Library) and MATLAB.
• Undertook various courses on control theory, electronics and fundamentals of robotics.

National Silver Medal

Sarajevo University

Recipient for outstanding academic results throughout Bachelor studies, with a grade point average of 9.31 out of 10.0.

STEM Games - Group 5th Placement (Technology Arena)

International Science, Technology, Engineering and Maths competition

Competed under the Faculty of Electrical Engineering Sarajevo by taking part in coding challenges in Poreč, Croatia.

Elektrijada - Bronze Group Medal (Engineering Mathematics 1)

International Electrical Engineering competition

Also achieved a group participation prize at the Technical English language competition in Budva, Montenegro.

Project Olympiad - Team 5th Place (Engineering Section)

International Burch University

Awarded for the electromechanical device project termed ‘Hydrogen core reactor’.

Machine-Learning Based Cache Miss Prediction

Springer International Journal on Software Tools for Technology Transfer - STTT

Published in Germany.

A Conformal Prediction-Based Framework for CPU Load Forecasting: A Black-Box Approach

COMPSAC 2025 Conference

Published in Toronto, Canada.

Abstraction-based Reduction of Input Size for Neural Networks

AISoLa 2023 Conference

Published in Greece.