I am Gerrit Luimstra, a Data Scientist and PhD student at the University of Groningen. My research and professional interests lie in efficient computer vision, machine learning, and AI systems that create positive societal impact.
Education
- PhD (part-time), University of Groningen (2025–present) – Research in efficient computer vision
- MSc Data Science and Systems Complexity, University of Groningen (2021–2024, Cum Laude, GPA 8.6/10)
- BSc Computing Science, University of Groningen (2018–2021, GPA 8.4/10)
- BSc Information Technology (propedeuse), NHL Stenden (2017–2018, Cum Laude)
Experience
Data Scientist, Municipality of Emmen (2022–present)
- Developing deep learning models for aerial imagery and real-time pedestrian counting.
- Implementing predictive models to support social policy.
- Building AI agents and platforms with OpenWebUI, Azure OpenAI, and ChromaDB.
- Integrating AI in RPA workflows and providing AI workshops within the organization.
Data Scientist, Innovatiespotter (2019)
- Developed a model with CBS to classify 500,000+ companies, identifying innovative firms in green energy.
Publications
- Luimstra, G., & Bunte, K. (2025). Solar Panel Segmentation on Aerial Images using Color and Elevation Information. ESANN 2025.
- Luimstra, G., & Bunte, K. (2022). Adaptive Gabor Filters for Interpretable Color Texture Classification. ESANN 2022.
- Luimstra, G., & Verschoore, W. Is it not yet Time to Swish? Comparing the ReLU and Swish Activation Functions. SC@RUG 2021–2022.
Achievements
- 1st place (shared) in the Acute Myeloid Leukemia Prediction Challenge
- Best Research Award (SC@RUG) for research on activation functions
- Top 3 placements in bachelor symposium and GAPC programming competition