Developing scientific software, research data infrastructure and interactive computational tools. Focus areas include high-performance spatial processing, scientific visualization, ecological telemetry, computer vision and reproducible researh workflows.
Scientific data infrastructure and interactive tools for ecological research and animal movement telemetry.
- EcoEngine Core — Java/Spring Boot backend for REST APIs, metadata ingestion and PostgreSQL data management.
- Ecosystem Frontend Suite — Angular/Nx monorepo integrating research applications for metadata management, dataset exploration, telemetry visualization and data curation.
- Eco-Proxy — Cloudflare Worker integration layer for Movebank API access and AI-assisted services.
Live applications: Specification Builder · Metadata Harvester · Dataset Explorer · Telemetry Animator · Sensor Streams · Spatial Visualizer · Tracking Curation
Technologies: Java, Spring Boot, PostgreSQL, Docker, TypeScript, Angular, Nx, Leaflet, Web Workers.
Interactive 3D point-cloud processing and visualization with spatial downsampling and non-destructive filtering. A C++20 computational engine compiled to WebAssembly and integrated with an Angular and Three.js frontend.
Technologies: C++20, WebAssembly, Emscripten, Angular, Three.js.
Scientific application for biomedical image inspection and processing, with deep-learning-based analysis and Grad-CAM visualizations to examine model attention.
Technologies: Python, PyTorch, FastAPI, computer vision, explainable AI.
A computer-vision project originating from a Master of Science project in 2010, with a fully renewed codebase in 2026. The current implementation combines a C++20 processing engine compiled to WebAssembly with a modern Angular interface for real-time facial analysis and visual exploration.
The project connects an earlier academic research interest with a contemporary browser-based implementation.
Technologies: C++20, WebAssembly, Emscripten, Ninja, Angular, TypeScript.
Browser-based application for processing facial blendshape data frame by frame and exploring temporal synchronization through an interactive timeline.
Technologies: Angular, TypeScript, MediaPipe, ECharts.
- Scientific computing: C++20, WebAssembly, computational processing, spatial data and performance-conscious implementations.
- Research data infrastructure: REST APIs, metadata ingestion, relational databases and data integration.
- Scientific visualization: 3D rendering, geospatial mapping, temporal analysis and interactive data exploration.
- Computer vision and AI: PyTorch, image analysis, model visualization and facial data processing.
- Software architecture: Java/Spring Boot, Python/FastAPI, TypeScript, Angular and Nx monorepos.
- Deployment and tooling: Docker, Git, Emscripten, Ninja and cloud-based application deployment.
The portfolio aims to support software that is understandable, maintainable and reusable in research contexts.
Key engineering priorities include:
- Clear documentation of architecture, algorithms and data flows.
- Automated testing and validation appropriate to each computational component.
- Reproducible development and deployment instructions.
- Transparent data provenance, methodological limitations and attribution.
- Modular interfaces and reusable components for scientific workflows.
Applied wellness and interactive reflection tools
- Bilateral Stimulation — Visual bilateral-stimulation tool.
- Breathe & Focus — Guided breathing and visual focus.
- Soundscape Mixer — Customizable layered audio environments.
- The Thought Shredder — Interactive cognitive-defusion exercise.
- 5-4-3-2-1 Grounding — Guided sensory-grounding exercise.
- Interactive Emotion Wheel — Interactive emotional vocabulary and self-reflection.
Open-source projects and live demonstrations are provided for technical exploration. Scientific validity, clinical efficacy and suitability for specific research workflows depend on the methods, data and validation documented by each project.