PecuHealth
AI-powered dairy health platform predicting mastitis risk and optimizing herd productivity through unified farm data.
MSc Artificial Intelligence • Utrecht University
Engineering solutions the world is missing
I'm an AI researcher and software engineer specializing in computer vision, deep learning, and graphics programming. My work bridges academic research with practical engineering, from developing neural networks for remote physiological monitoring to building custom game engines in DirectX and OpenGL.
Currently exploring procedural 3D generation and co-founding PecuHealth and ClusTEQ. My research has been published in academic venues, focusing on contactless health monitoring, NLP bias detection, and ethical AI systems.
Co-founder driving AI products in health-tech and BIM innovation
AI-powered dairy health platform predicting mastitis risk and optimizing herd productivity through unified farm data.
Next-generation BIM platform built on game-engine tech, delivering lag-free design, real-time collaboration, AI assistance, and advanced versioning.
A selection of academic research and engineering work
Master's thesis project: Deep learning approach for contactless heart rate monitoring via facial video analysis. End-to-end neural network for remote photoplethysmography.
Real-time face detection and tracking system for identity management. Implements state-of-the-art face recognition algorithms with efficient processing pipeline.
Content-based multimedia search engine using feature extraction and similarity metrics. Implements advanced indexing and query optimization techniques.
48-hour collaborative software challenge building sign language recognition from video, with rapid prototyping and agile development under tight time constraints.
Research contributions in AI, computer vision, and ethical computing
Utrecht University, November 2021
Deep learning approach to remote heart rate estimation using spatial-temporal representations. Novel methodology for contactless cardiovascular monitoring through computer vision, eliminating the need for traditional touch sensors.
Tallinn University of Technology, February 2021
Novel approach to language bias visualization through temporal analysis of large-scale Reddit datasets. Introduces methodology for collecting and analyzing bias scores from text corpora using word embeddings.
April 2020
Examines decentralized systems as solutions to privacy policy challenges. Presents ethical dilemmas in modern computing and proposes decentralized architectures as ethically supportive alternatives.
Recent personal work and GitHub activity
Technical expertise across AI, graphics programming, and software engineering
Let's connect and discuss opportunities