Amsterdam Area
Experienced Machine Learning Engineer with a demonstrated history of applying AI technologies in various industries. Successfully contributed to several teams, ranging from small scale startups to bigger multinational corporations. Strong computer science and engineering professional with a Master of Science focused in Artificial Intelligence from University of Amsterdam. My skill set in Computer Vision and Deep Learning makes me a right fit for both research and/or application oriented projects.
At Nicolab, we believe connecting human and artificial intelligence will revolutionize emergency care. Founded in 2015, Nicolab is a MedTech company that stems from leading clinical research. We develop end-to-end solutions to further empower physicians in emergency care. Powered by unique datasets, our artificial intelligence product StrokeViewer® enhances stroke patient outcomes by reducing time to treatment.
Co-author of multiple machine learning publications, focusing on computer vision topics for innovative and applicable use-cases, such as: • Preserving structural information with adversarial learning for near real-time applications (utilized in Shell Autonomous Integrity Recognition application) • Minimizing performance disparity (focusing on rare classes) and promoting fairness in semantic segmentation • Improving (cross-)user performance by federated learning for non-IID data in computer vision
Worked on the object detection part of Shell Autonomous Integrity Recognition: • Real-time detection of external corrosion and insulation issues to reduce leakages in equipment (such as meter piping, flanges and dead ends) • Reduce cost of maintenance • Improve the quality, efficiency and standardization of visual inspections • Powered by the BHC3™ AI Suite and Microsoft Azure.
• Designed and implemented computer vision and deep learning pipeline for detecting and tracking pigs’ activities and illnesses. • Maintained cloud computing and data storage system on AWS for computer vision projects. • Created object detection datasets through processing video streams from farms. • Developed a video annotation tool for labeling behavior timelines. • Communicated with partners and clients about product installation. • Software stack: Python, Pytorch, Tensorflow, OpenCV.
• Research and development of artificial neural architectures for unmanned aerial vehicles (UAV) as part of the Sense and Avoid project. • With my research, I achieved a 2nd place in Signal processing at BUTE's Student Paper Competition, 2017. • Software stack: Python, Keras.
• Created prototypes of efficient neural models for drones. • Played around with tools for understanding and visualizing neural networks.
• Researched AI technologies for CRM Systems. • Conducted several experiments with AWS Machine Learning Services.