I am an AI PhD Candidate and engineer with a passion for turning complex problems into impactful solutions. My journey began during my bachelor's degree, where I tackled my first AI project—movement recognition from EEG signals—despite having no formal AI coursework. From those early experiments, I've continued to push my boundaries across academia, entrepreneurship, and applied research.

During my master's studies, I explored a wide range of AI applications: from traffic sign recognition with both traditional ML and deep learning, to resource-constrained activity recognition systems, and a university-focused chatbot built before modern LLMs became mainstream. In parallel, I co-led the development of Unicas2Go, a full-stack smart carpooling app that reached over 1,000 users in the University of Cassino community, combining technical development with entrepreneurial execution.

Currently, I am pursuing a PhD on AI for Alzheimer's disease diagnosis and prognosis using 3D MRIs. I've developed explainable "2.5D" attention-based models for scarce-data scenarios, diffusion-based generative methods for unsupervised learning, and large-scale training pipelines across multi-GPU clusters. My work has been validated by clinical experts and expanded through collaborations with leading research centers in medical imaging.

My long-term vision is to contribute to the development of generative foundation models bridging the gap between advanced AI research and real-world impact.