University of Zagreb, Faculty of Electrical Engineering and Computing,
Zagreb, Croatia
Master of Science in Electrical Engineering and Information Technology,
specializing in Signal Processing and Control Theory, 2015
Machine Learning Engineer with 9+ years of experience in applied deep learning, computer vision, and AI-driven industrial systems. Proven track record in developing production-level ML pipelines, working with cross-functional teams, and delivering intelligent systems for domains including transportation, industrial processes, hydrology, and radar sensing.
Languages: Python, C++, C, MATLAB
Machine Learning: CNN, RNN (LSTM, GRU), GNN, Attention,
Reinforcement Learning, Symbolic Regression, Evolutionary Algorithms,
Boosting, SVM
Frameworks & Libraries: PyTorch, TensorFlow, Pytorch Lightning,
Scikit-learn, NumPy, SciPy, OpenCV, ONNX, FastAPI
Embedded / Edge AI: CUDA, Darknet (C), TensorRT
Data Engineering: Pandas, SQL, Data Pipelines
Cloud & DevOps: AWS, Docker, Git, Jira
Geolux Machine Learning Engineer
Jan 2025 -- Present
Designed and deployed ML algorithms on edge radar sensors for real-time estimation of water velocity, snow water equivalent, and ice presence.
Developed lightweight models optimized for constrained embedded environments, balancing accuracy with latency and power efficiency.
Collaborated with hardware engineers to integrate ML solutions directly into sensor firmware, enabling intelligent environmental monitoring in the field.
Glass Service / Saint-Gobain SEFPRO Machine Learning Engineer
Apr 2020 -- Jan 2025
Developed a deep learning based segmentation model for visual inspection and control in high-temperature glass production environments.
Designed ML-driven control systems using model predictive control (MPC) and reinforcement learning (RL), enabling adaptive process optimization.
Tackled complex, high-dimensional industrial systems where traditional control approaches failed due to numerous disturbance variables and nonlinear dependencies.
Telegra Machine Learning Engineer
Jun 2017 -- Mar 2020
Developed deep learning models for real-time traffic monitoring and classification in intelligent transportation systems.
Designed detection systems to identify and raise alarms for congestion, debris, animals, pedestrians, and vehicle collisions, enhancing road safety and operational response.
GlobalLogic Software Development Engineer
Nov 2015 -- Jun 2017
Maintained and extended software components for dSPACE platforms, enabling hardware-in-the-loop (HIL) simulation for automotive system development and validation.
Worked primarily in MATLAB and C++ to troubleshoot, adapt, and support simulation configurations.
LABUST (Laboratory For Underwater Systems and Technologies) External Associate
Jan 2015 -- Nov 2015