I build data-intensive systems at the intersection of data engineering, machine learning, and software architecture. The focus is production-grade work: messy data, latency, scale, reliability, and systems that remain understandable after the demo.
- Background: EngD in Data Science
- Core interests: streaming platforms, digital twins, spatiotemporal systems, applied AI
- Default ecosystem: Python
My work is implementation-driven and oriented toward operational relevance. Recent work includes an Urban Digital Twin for the City of ’s-Hertogenbosch, integrating:
- real-time streaming pipelines
- geospatial and spatiotemporal data processing
- time-series storage and analytics
- forecasting and machine-learning models
- interactive 3D visualization
This work has been presented in academic and professional venues and continues to evolve into deployable systems.
- real-time and streaming analytics
- data integration and interoperability
- federated and distributed data processing
- ML deployment, evaluation, and monitoring in production
- system reliability, scalability, and reproducibility
I prefer systems that are:
- composable (replaceable parts, not fragile monoliths)
- explicit (clear interfaces and data contracts)
- inspectable (debuggable without ritual sacrifices)
- maintainable (designed for the second year, not the second week)
I am an open-source practitioner and contribute primarily within the Python ecosystem. I value clarity over cleverness, and reproducibility over mystery.
Through DataTwinLabs, I collaborate with public organizations and industry partners on data platforms, digital twins, and applied AI systems.
If you want to collaborate on research, engineering, or applied projects where data and real-world systems meet, reach out.
- Website:
<datatwinlabs.nl> - LinkedIn:
<https://www.linkedin.com/in/danielwondyifraw/> - Publications / talks:
<https://www.jads.nl/news/paving-the-way-for-sustainable-urban-construction/> - Contact:
<datatengineerd[at]outlook[dot]com>