Bob Giezi developed end-to-end Jupyter notebooks for the IBM/terratorch repository, focusing on generating embedding vectors from Sentinel-2 RGB imagery using the TerraMind deep learning model. He designed reproducible workflows that integrate geospatial data with machine learning, enabling rapid experimentation and smoother onboarding for new users. His work included model setup, image processing, and embedding retrieval, all implemented in Python with PyTorch and GIS tools. By providing ready-to-run templates and practical demonstrations, Bob addressed business needs for client engagement and reproducibility. The depth of his contribution lies in delivering practical, production-ready resources that streamline geospatial deep learning experimentation.