🧭 About Me

I’m Rajiv Ranjan, a geospatial AI researcher building practical, data‑driven tools for agriculture and climate resilience. My work blends satellite and UAV imagery with self‑supervised learning to extract reliable signals stress, yield potential, phenology without heavy labels. I design lightweight spectral models, agentic AI workflows, and interactive dashboards so insights move from notebook to field quickly. Recently, I’ve been prototyping pixel‑wise spectral MAE variants, agentic research assistants, and Hugging Face–hosted portfolio tools to document and share reproducible science.

During my time at Purdue, I aim to focus on developing and evaluating multi-scale, multi-source data fusion techniques using both satellite and UAV data for sugarcane growth monitoring. Additionally, I plan to pretrain a Geospatial Foundation Model leveraging various pretraining paradigms including SSL, multi-objective learning, and multimodal fusion models, followed by fine-tuning on labeled datasets

My Experiment's Garbage