I'm a machine learning engineer working on computer vision, geospatial ML and data engineering. I have a PhD in physics and computational neuroscience.

What I work on

• Satellite imagery. I train segmentation models (U-Net, SegFormer) for surface classification. I then export them to ONNX so they run inside a client's commercial .NET software.
• OSINT and investigations. I build data platforms on Google Cloud that collect, parse and OCR open-source material, so investigators can search it and connect records. Before that, I developed red-flag indicators for EU beneficial-ownership data.
• Self-organizing neural networks. My PhD at Goethe University Frankfurt (FIAS and the Max Planck Institute for Brain Research) studied how criticality and learning emerge in recurrent networks.

Most of this lives in private client repositories. Here's what's public:

• SORN_V2: a self-organizing recurrent neural network in Python 3, with experiments from my PhD
• sandpilemodel: the Bak–Tang–Wiesenfeld sandpile and its power-law avalanches

Now: I'm building small projects with open data, both science and OSINT, each with interactive visualizations.

Publication: Del Papa, Priesemann & Triesch (2017). Criticality meets learning: Criticality signatures in a self-organizing recurrent neural network. PLOS ONE.

Tools: Python · PyTorch · scikit-learn · OpenCV · GCP (BigQuery, Cloud Run, Document AI) · SQL · QGIS · JavaScript · C# / .NET · ONNX · Docker

LinkedIn · Lisbon, Portugal