About Me
I am an Assistant Professor at Old Dominion University with a joint appointment in the Department of Computer Science and the School of Cybersecurity.
My research focuses on the security, robustness, and applicability of Graph Neural Networks (GNNs) and Large Language Models (LLMs).
My goal is to advance AI security technologies that enhance trust and reliability in the systems we depend on today and in the future. Overall, I have contributed to more than 20 conference papers, over 13 journal papers, 10 US and international patents and patent applications, and 1 book chapter, and have lectured computer engineering and related courses at NJIT and in universities in Turkey and the UAE.
Research Interests
- Currently: Security and robustness of GNNs and LLMs
- Adversarial robustness and prompt optimization
- Applications in secure and functional source code generation
- Hardware design automation with LLMs and GNNs
- Transportation system analytics and Internet security
- Deepfake detection using multimodal and graph-based methods
- Previously: Machine learning for communications (e.g., physical layer security, channel estimation, spectrum sensing)
Education
Ph.D. in Computer Engineering
NJIT, 2021 – 2025
M.Sc. in Electrical and Electronic Engineering
Eastern Mediterranean University, 2010
B.Sc. in Electrical Engineering
Birzeit University, 2009
Research Interests Overview
This visualization summarizes my research areas, from past work in machine learning for communications to current interests in AI security and robustness, including secure code generation, deepfake detection, and adversarial attacks.
Contribution Map
This contribution map illustrates the areas of my research across software engineering, adversarial machine learning, hardware design automation, predictive modeling, and real-world ML applications. It also highlights top-tier venues like CCS, IEEE S&P, GLSVLSI, and others.
Teaching at Old Dominion University
Graduate Courses (M.S./Ph.D.)
-
CS 795/895 – Special Topics in Computer Science: Large Language Model (LLM) Architectures and Applications
(Spring 2026)
This graduate special topics course provides an in-depth exploration of the architectures, training paradigms, and applications of modern large language models. Starting from foundational ML and NLP concepts, the course progresses to transformer-based LLM architectures, prompt optimization, and retrieval-augmented generation (RAG). It also examines multimodal extensions via vision-language models (VLMs) and the emerging role of LLMs as core reasoning engines in agentic AI systems capable of planning and tool use. Students study pre-training objectives, fine-tuning strategies, and state-of-the-art evaluation metrics, and are prepared to analyze, design, and apply LLM-based systems in both research and real-world contexts.
Syllabus (PDF) -
CS 722/822 – Machine Learning
(Fall 2025)
This graduate course introduces core machine learning methods, including supervised and unsupervised learning, regression, classification, clustering, anomaly detection, and dimensionality reduction, with emphasis on evaluation, cross-validation, calibration, and error analysis.
Syllabus (PDF)
Professional Experience
-
Assistant Professor, Department of Computer Science & School of Cybersecurity, Old Dominion University, Norfolk, VA, USA (2025 – Present)
- Tenure-track faculty appointment focusing on AI security, robust machine learning, and applications of GNNs and LLMs.
-
Research Assistant / Teaching Assistant, New Jersey Institute of Technology, Newark, NJ, USA (2021 – 2025)
- Research on LLM and GNN security and applications, including secure code generation and adversarial robustness.
- Lectured undergraduate courses in computer engineering.
-
Lecturer, Abu Dhabi Vocational Education and Training Institute, Abu Dhabi, UAE (2019 – 2021)
- Taught undergraduate electrical engineering courses.
- Supervised student projects.
-
Lecturer, Izmir University of Economics, Izmir, Turkey (2016 – 2017)
- Taught courses in electrical engineering.
- Guided graduation projects.