The purpose of this research is to present EcoHeno 1.0, a web and mobile application designed to optimize hay production and advance toward digitally interconnected sustainable systems in Colombian livestock farms. This proposal responds to the need to reduce losses, improve traceability, and facilitate decision-making in rural contexts with limited connectivity. The methodology adopted an applied-experimental approach, combining participatory diagnostics with iterative-incremental design under agile Scrum and Domain Driven Design principles. It integrated qualitative techniques (interviews and observation), computational modeling, and quantitative validation through artificial intelligence. The application was developed with lightweight and scalable technologies such as React Native, SQLite-Cloud, and scikit-learn. As a proof of concept, multiple regression and decision tree (CART) models were trained, alongside more advanced algorithms such as Random Forest and Gradient Boosting. Pilot testing on livestock farms demonstrated a 15% reduction in annual hay losses, a 25% increase in productivity, and a 90% decrease in data entry errors. Validation with R2 and MSE metrics showed that Random Forest and Gradient Boosting significantly outperformed linear models, achieving more accurate and robust predictions. EcoHeno 1.0 represents an innovative contribution to digital agriculture, demonstrating that advanced computational models can be successfully applied to optimize agro-productive processes, even in rural environments facing technical and infrastructural limitations