A curated selection reflecting my cybersecurity interests, research direction, and technical execution.
Detecting package hallucinations in LLM-generated code suggestions and benchmarking behavior across GPT-4, Gemini, and Cohere. Evaluates the reliability of AI-generated code recommendations by identifying fictional or non-existent software packages.
View detailsIdentifying and preventing unwanted or suspicious swipe-style interactions, with focus on detection logic and secure interaction patterns. Explores gesture-based security vulnerabilities and implements protective measures.
View detailsA recommendation system exploring intelligent content suggestions through data-driven modeling and user preference patterns. Implements collaborative filtering and content-based approaches for personalized recommendations.
View detailsA biometric web login flow exploring face recognition, identity verification, and secure access patterns. Integrates computer vision libraries for real-time facial detection and matching with liveness detection.
View detailsA hands-on malware analysis environment for studying suspicious files, observing behavior, and documenting technical indicators. Includes sandboxed execution, static analysis, and dynamic behavior monitoring.
View detailsStreamlines attendance tracking through applied software development and automation logic. Features real-time tracking, automated reporting, and integration with existing institutional systems.
View detailsMy graduate research at the NYU Osiris Lab explored the intersection of artificial intelligence and cybersecurity.
My research focused on understanding how adversarial techniques can compromise machine learning models, with particular emphasis on LLM supply chain vulnerabilities. The Package Hallucination Detection System serves as a foundational tool for identifying when LLMs suggest non-existent dependencies — a vector that could lead to supply chain attacks if malicious actors register these hallucinated package names.
View details