▪ Developed and implemented an advanced autoencoder model for anomaly detection in Rocky Mountain Elk, utilizing a custom loss function to enhance model sensitivity and accuracy. This model was designed to prioritize significant anomalies, crucial for the early detection of Chronic Wasting Disease (CWD).▪ Significantly enhanced the YOLOv5 model for wildlife monitoring, achieving a precision of 91.7% and a recall of 68% across all classes, with an overall mAP50 of 75.1%. The model excelled particularly in detecting Tule Elk, demonstrating a precision of 93.9%, recall of 94.2%, and mAP50 of 94.6%, showcasing its effectiveness in varied environmental conditions using geospatial satellite imagery.▪ Established collaborative research with Dr. DeVivo (Washington Department of Fish and Wildlife), Dr. Edmunds (Colorado State University), and Dr. Street (Mississippi State University), integrating interdisciplinary expertise to refine disease detection methods through AI. This collaboration is directed towards a joint publication, emphasizing the project’s innovation and impact.▪ Spearheaded the analysis and processing of large datasets, ensuring robust model training and validation, which contributed to groundbreaking insights into wildlife health monitoring.

▪ Played a key role in data collection preprocessing, and analysis significantly enchaining data handling and providing robust support for multiple AI projects.▪ Fostering a strong partnership with the CEO & CTO, gaining invaluable insights, and acquiring industry-specific skills critical to AI engineering.▪ Contributed to the optimization and refinement of existing algorithms, developing a comprehensive understanding of algorithmic efficiency and performance.

▪ Collaborated with the Director of University of California research center to evaluate citrus responses to a foliar spray enhancing growth in low nitrogen conditions. ▪ Spearheaded data processing and microbial community identification in the rhizosphere of citrus plants:o Successfully analyzed 16S and ITS data using Qiime2, DADA2, Greengenes2, and Ampviz2.o Identified key endophytes promoting salt tolerance, and recognized, known plant-growth-promoting microbes. ▪ Applied AI / Machine Learning techniques to relate to microbe datasets to environmental conditions:o Implemented Random Forest analysis, Gradient Boosting Classifier, Logistic Regression, and SVC.o Used GridSearchCV for hyperparameters to predict significant biomarkers related to foliar spray and fertilizer conditions. ▪ Research findings are in the process of being published at the end of 2024.