Migrated to a more robust model architecture using Gradient Boosting Machines (GBM) via XGBoost, enabling multi-variable predictions and control. AI now autonomously manages irrigation, fertilization, and fan systems, while flagging health anomalies using image-based inference.

Upgraded the system with soil moisture, CO₂, and light intensity sensors to improve data granularity. Built a custom web-based dashboard to visualize environmental variables and plant health, significantly enhancing monitoring and system responsiveness.

Initiated the project by installing a Raspberry Pi 5 in a home greenhouse environment. Integrated Adafruit DHT22 sensors to collect temperature and humidity data, forming a foundational dataset tailored to optimizing the propagation of Juniperus Itoigawa cuttings.