I am Jin Wang, a PhD Candidate in the Animal Sciences Department at the University of Florida, specializing in AI for agriculture: computer vision, machine learning, and scalable data pipelines for precision livestock farming. I completed both my B.S. and M.S. in the Electrical and Computer Engineering department, which gives me a strong foundation in building reliable, production-oriented systems. My experience includes developing and deploying an automated on-farm phenotyping pipeline that integrates sensing and AI on edge devices, enabling real-time BW/BCS prediction while also organizing high-throughput data capture for downstream analysis. I am the first author of ShinyAnimalCV, an open-source, cloud-based web application for object detection, segmentation, and 3D visualization, built to make model outputs and data quality issues interpretable for users. For model training and experimentation at scale, I regularly use UF’s HiPerGator HPC cluster as a remote compute environment to run deep learning workloads, manage datasets, and reproduce experiments across different configurations. In addition, I have published work on time-series modeling using tri-axial accelerometer data and on rigorous evaluation design, and I’ve recently evaluated transfer learning strategies on multi-farm datasets to improve performance when labeled data are limited and conditions vary. What makes me stand out is a data-centric mindset and strong “build + validate + iterate” execution: I’m comfortable combining rule-based logic with ML, collaborating closely with domain experts, and shipping reliable, user-facing intelligence. I primarily work in Python (TensorFlow/PyTorch) and can support cloud-ready backends and APIs as the product evolves.…I am Jin Wang, a PhD Candidate in the Animal Sciences Department at the University of Florida, specializing in AI for agriculture: computer vision, machine learning, and scalable data pipelines for precision livestock farming. I completed both my B.S. and M.S. in the Electrical and Computer Engineering department, which gives me a strong foundation in building reliable, production-oriented systems. My experience includes developing and deploying an automated on-farm phenotyping pipeline that integrates sensing and AI on edge devices, enabling real-time BW/BCS prediction while also organizing high-throughput data capture for downstream analysis. I am the first author of ShinyAnimalCV, an open-source, cloud-based web application for object detection, segmentation, and 3D visualization, built to make model outputs and data quality issues interpretable for users. For model training and experimentation at scale, I regularly use UF’s HiPerGator HPC cluster as a remote compute environment to run deep learning workloads, manage datasets, and reproduce experiments across different configurations. In addition, I have published work on time-series modeling using tri-axial accelerometer data and on rigorous evaluation design, and I’ve recently evaluated transfer learning strategies on multi-farm datasets to improve performance when labeled data are limited and conditions vary. What makes me stand out is a data-centric mindset and strong “build + validate + iterate” execution: I’m comfortable combining rule-based logic with ML, collaborating closely with domain experts, and shipping reliable, user-facing intelligence. I primarily work in Python (TensorFlow/PyTorch) and can support cloud-ready backends and APIs as the product evolves.WWWW…