I design AI-driven systems for cybersecurity training, blending reinforcement learning, intelligent tutoring,and adaptive VR to build safer, smarter learning experiences.
Tools & domains I work with: Python, PyTorch, Unity (VR), SwiftUI, RL pipelines, RAG/guardrails, and secure deployment.
The Adaptive Content Engine (ACE) is a reinforcement learning–driven VR training system that personalizes cybersecurity education. It uses PPO and AIRL models to dynamically adjust challenge difficulty based on player performance, improving both engagement and learning efficiency.
The AI Tutor project applies reinforcement learning–guided chain-of-thought (CoT) to enhance the reasoning and adaptability of large language models in tutoring contexts. It integrates retrieval-augmented generation (RAG) and safety guardrails to deliver secure, context-aware, and citation-first feedback for cybersecurity and AI education.
