Terra, CSU × USDA × NASA Geospatial AI
Won the 24-hour overnight CSU Geospatial AI Hackathon (USDA, NASA, Databricks, Esri, AWS) with a two-track ML system.
Corn yield forecasting joins six independent sources into a 26,924-row Parquet across five states and twenty years, with per-forecast-date XGBoost regressors at +46.7% RMSE lift over baseline, calibrated uncertainty cones, and a four-tool Claude agent writing cited plain-English reports.
Land-cover and GHG pairs a SegFormer-B1 model fine-tuned on LoveDA (val mIoU 0.5147) with an IPCC and EPA cited emissions calculator and a "what-if" land-use agent.
State-level forecasts land within as little as 0.6% of the USDA ground truth, with the best counties under 0.3% error, a +46.7% RMSE gain over baseline.
