Maroje Marohnic
Verified Expert in Engineering
Machine Learning Developer
Maroje is a machine learning engineer with a PhD in applied mathematics plus five years of experience developing computer vision applications for autonomous driving and machine vision systems. Passionate about artificial intelligence and software engineering, Maroje is looking for projects that allow him to create new software and machine learning solutions showing his proficiency in deep learning algorithms, TensorFlow, PyTorch, C, C++, Python, MATLAB, and embedded systems.
Portfolio
Experience
Availability
Preferred Environment
Linux, Python, C, C++, TensorFlow
The most amazing...
...thing I've developed is an anomaly detector on a manufacturing line. It was trained on only 100 images and worked perfectly well.
Work Experience
Lead Machine Learning Engineer
Ascalia
- Developed machine vision applications used in manufacturing plants related to image segmentation and anomaly detection.
- Selected the machine vision system components: lighting, camera, and lenses. Also, managed data collection and marking processes, trained and evaluated algorithms, handled deployment, and prepared presentations for customers.
- Implemented machine learning best practices in the company–data collection, data annotation, and AWS services for training.
- Led a group of four engineers, quality control, and served as a mentor for two projects.
Senior Research and Development Engineer, Team Lead
Visage Technologies
- Worked on advanced driver-assistance systems L2 and L3 based on computer vision for Veoneer and Arriver. Progressed from mid to senior engineer and took a team lead role in three years.
- Developed computer vision applications; wrote requirements and data collection instructions, modeled CNN architectures, ran and monitored training, improved the training process, and deployed the applications.
- Wrote target code in C and C++ following MISRA C standards.
- Created a profiling tool to analyze the system's memory usage.
- Led and onboarded a new team of four developers, supporting them through various tasks such as optimizing statistical tools, generating new stereo measurements, tuning tracker, and deploying functional safety jobs.
Software Engineer
GlobalLogic
- Developed and maintained dSPACE GmbH's real-time simulation software for the automotive industry.
- Analyzed and optimized the simulation engine for embedded systems, handled processes and multiprocessor platforms tasks, and held the kernel maintenance.
- Improved the initialization of the models significantly.
Postdoctoral Researcher and Teaching Assistant
University of Zagreb
- Awarded the best young mathematician for research achievements in 2015.
- Developed research in applied mathematics, including partial differential equations, theory of elasticity, calculus of variations, and theory of thin elastic bodies.
- Taught in the Mathematics course optimization, discrete mathematics, probability and statistics, ordinary differential equations, methods of mathematical physics, numerical mathematics, and fundamentals of algorithms.
Experience
Developing Machine Vision Applications
Skills
Languages
Python, C, SQL, Simulink, R, C++
Libraries/APIs
TensorFlow, XGBoost, OpenCV, Keras, NumPy, PyTorch
Tools
Git, Amazon SageMaker, Boto 3, MATLAB, Jenkins, DOORS
Platforms
Amazon Web Services (AWS), Linux, Docker, NVIDIA CUDA
Storage
Amazon S3 (AWS S3), MySQL
Other
Research, Mathematics, Programming, Computer Vision, Deep Learning, Neural Networks, Convolutional Neural Networks (CNN), Machine Vision, Autonomous Navigation, Software Development, Machine Learning, Datasets, Artificial Intelligence (AI), Deep Neural Networks, Image Processing, Object Detection, Deep Reinforcement Learning, University Teaching, Dashboards, Natural Language Processing (NLP), GPT, Generative Pre-trained Transformers (GPT)
Paradigms
Data Science
Education
PhD in Applied Mathematics
University of Zagreb - Zagreb, Croatia
Master's Degree in Mathematics
University of Zagreb - Zagreb, Croatia
Certifications
Sequence Models
Coursera
Deep Reinforcement Learning Nanodegree Program
Udacity
Deep Learning Nanodegree Program
Udacity
How to Work with Toptal
Toptal matches you directly with global industry experts from our network in hours—not weeks or months.
Share your needs
Choose your talent
Start your risk-free talent trial
Top talent is in high demand.
Start hiring