PhD Candidate, The City College of New York, United States
Ejup Hoxha is a Machine Learning Engineer at Amazon Web Services (AWS) in New York, specializing in Large Language Models (LLMs), time series forecasting, and machine learning/deep learning. He is also a PhD candidate in Electrical Engineering at The City College of New York. With experience in robotics, sensor fusion, visual SLAM, and computer vision, Ejup contributes significantly to the field of non-destructive testing (NDT). His work spans across robotics, automation, and software development, making him a versatile and innovative engineer. Ejup has contributed to multiple research projects and has served as an adjunct lecturer, teaching courses related to robotics and engineering. His research, aimed at improving construction and infrastructure processes, has earned recognition in prestigious journals and conferences.
Professional Profile
Google Scholar
Scopus
Strengths for the Award
Ejup Hoxha has demonstrated an exceptional ability to merge cutting-edge machine learning and robotics technologies with practical applications in infrastructure inspection, particularly in non-destructive testing (NDT) and robotic systems. His contributions to fields like Ground Penetrating Radar (GPR) imaging, robotic inspection, and subsurface defect mapping are groundbreaking, as evidenced by his high-quality publications and their citations in top-tier journals and conferences. Notably, his work on automated GPR reconstruction and impact-echo methods for concrete inspection is both innovative and impactful, addressing real-world challenges in construction and infrastructure maintenance. His strong expertise in robotics, reinforcement learning (RL), and sensor fusion enhances his ability to propose novel solutions in both academic and industrial settings. Furthermore, his leadership in developing secure and scalable systems at AWS further solidifies his role as a pioneering researcher.
Areas for Improvements
While Ejup has achieved great success in his technical work, expanding his focus to the commercial viability and broader industrial applications of his research could make his innovations even more impactful. His future work could benefit from fostering collaborations with multidisciplinary teams to integrate more cross-sector knowledge, which would help create versatile and adaptable systems that address a broader range of industry needs. Additionally, increasing his outreach and visibility through more industry-driven projects, real-world implementation, and knowledge-sharing platforms could further enhance the practical application and adoption of his work.
Education
Ejup Hoxha is currently pursuing a PhD in Electrical Engineering at The City College of New York. He holds a Master of Philosophy in Electrical Engineering (2023) from the same institution. Ejup completed a Master of Science in Computer Engineering (2020) and a Master of Science in Computerized Automation and Robotics from the University of Pristina in Kosovo. His educational foundation began with a Bachelor of Science in Electrical and Computer Engineering, specializing in Automation, from the University of Pristina in 2015. Ejup’s rigorous academic background supports his expertise in machine learning, robotics, and control systems, enabling him to lead innovative research and practical applications in his field.
Experience
Ejup Hoxha currently works as a Machine Learning Engineer II at AWS, where he specializes in developing automated LLM evaluation methods and fine-tuning systems. Prior to this, he was a Software Development Engineer I at AWS, responsible for designing secure, scalable, distributed systems. As a Graduate Research Assistant and Adjunct Lecturer at The City College of New York, Ejup led robotics projects involving reinforcement learning (RL), sensor fusion, visual SLAM, and computer vision. He has also worked as a Robotic Systems Engineer at InnovBot LLC, where he developed sensor fusion and control algorithms. Additionally, Ejup has experience in SCADA software development and automation, gained during his roles at N.P. INET and Call Home Electronics in Kosovo.
Awards and Honors
Ejup Hoxha has received recognition for his contributions to robotics and machine learning. His work in robotics, particularly in the area of robotic inspection and subsurface defect mapping, has been presented in renowned conferences and journals. He has been cited for his research on ground penetrating radar (GPR) and robotic systems for underground utilities. Ejup’s academic excellence has been acknowledged through multiple research awards, including his publication in IEEE Sensors Journal and the Journal of Computing in Civil Engineering. His achievements reflect his deep commitment to advancing robotics and NDT technologies, earning him the respect of peers in the engineering community.
Research Focus
Ejup Hoxha’s research focuses on the intersection of machine learning, robotics, and non-destructive testing (NDT). He specializes in robotic systems for infrastructure inspection, employing techniques like reinforcement learning, sensor fusion, and computer vision to enhance the efficiency of underground utility surveys and concrete inspections. His work with ground penetrating radar (GPR) and impact-echo methods aims to improve subsurface defect mapping and utility reconstruction. Additionally, Ejup’s research explores the application of artificial intelligence and deep learning to automation systems, with a focus on time-series forecasting and the development of automated LLM evaluation methods. His interdisciplinary research contributes to the evolution of smart systems for infrastructure monitoring and maintenance.
Publication Top Notes
- GPR-based model reconstruction system for underground utilities using GPRNet 📑
- Improving 3D Metric GPR Imaging Using Automated Data Collection and Learning-based Processing 📘
- Robotic inspection of underground utilities for construction survey using ground penetrating radar 📍
- Robotic Inspection and Subsurface Defect Mapping Using Impact-echo and Ground Penetrating Radar 🔧
- Robotic Inspection and Characterization of Subsurface Defects on Concrete Structures Using Impact Sounding 🏗️
- Automatic Impact-sounding Acoustic Inspection of Concrete Structure 🔊
- Robotic Inspection and 3D GPR-based Reconstruction for Underground Utilities 🛰️
- Contrastive learning for robust defect mapping in concrete slabs using impact echo 🎯
Conclusion
Ejup Hoxha is a deserving candidate for the Best Researcher Award. His innovative contributions to robotics, machine learning, and infrastructure inspection place him at the forefront of research in these fields. His ability to leverage advanced AI and robotics technologies to address challenges in non-destructive testing and construction is exemplary. With continued focus on collaboration and the commercialization of his work, Ejup has the potential to make an even greater impact on both academic and industrial domains. His research accomplishments, technical expertise, and commitment to advancing knowledge in his field make him an excellent contender for this prestigious award.