Background & experience
About my research
Current Position
I’m an AI researcher and full-stack software developer with a strong foundation in machine learning, generative AI, scientific computing, and signal processing. Currently, I serve as a Research Scientist at Texas A&M AgriLife Research, where I lead interdisciplinary projects at the intersection of AI, remote sensing, and life sciences.I have a proven track record of developing end-to-end machine learning pipelines—from data preprocessing and model design to distributed GPU training and deployment in both production and research environments. My expertise spans tools and frameworks such as TensorFlow, PyTorch, Kubeflow, Ray, OpenCV, Open3D, Docker, and Kubernetes.
My recent research focuses on leveraging the emerging capabilities of large language models (LLMs)—including reasoning, planning, and multi-agent collaboration—to streamline analytical workflows for multimodal remote sensing data. This includes integrating and processing ground-based field observations, UAV imagery, satellite data, LiDAR, and Ground Penetrating Radar (GPR) to support applications in scientific research automation and agricultural digital phenotyping. As part of this work, I’ve gained hands-on experience fine-tuning and aligning LLMs using advanced prompting techniques such as Chain-of-Thought (CoT) and ReAct, along with Parameter-Efficient Fine-Tuning (PEFT) methods like LoRA and Direct Preference Optimization (DPO). I’ve also implemented multimodal RAG systems and designed multi-agent workflows for complex reasoning tasks using tools such as ADK, CrewAI, Hugging Face, LangChain, AutoGen, LangGraph, and other generative AI frameworks.
My background in computer vision has also enabled me to contribute to multidisciplinary research efforts, developing tools for feature extraction, object detection, and image segmentation using CNNs, Vision Transformers (ViT), transfer learning, and semantic segmentation. I have applied these methods across multidimensional image datasets, from microscopy to hyperspectral imagery, enabling high-throughput, automated disease detection and lab optimization.
