Politecnico di Torino logo
Personale

Hadi Ibrahim

Dottorando in Ingegneria Informatica E Dei Sistemi , 41o ciclo (2025-2028)
Dipartimento di Automatica e Informatica (DAUIN)

Profilo

Dottorato di ricerca

Argomento di ricerca

Secure Artificial Intelligence: Enhancing IT Infrastructure and Online Services

Tutori

Keywords

Data science, Computer vision and AI

Biografia

I am Hadi Ibrahim, a PhD student at Politecnico di Torino. I hold a Laurea Magistrale in Data Science and Engineering from Politecnico di Torino and a Bachelor of Science in Computer Science from Rafik Hariri University in Lebanon. During my master's, I developed a strong research foundation spanning natural language processing, deep learning, computer vision, and generative AI, working on projects including LLM-based assistants with retrieval-augmented generation, abstractive dialogue summarization, and visual place recognition. This culminated in a thesis proposing an an extended CycleGAN architecture for multi-style and mixed-style text transfer, demonstrating competitive performance against large language models while requiring significantly fewer computational resources. The thesis is currently under evaluation and the work has been submitted for publication. Before starting my PhD, I worked as a Data Scientist applying reinforcement learning, multi-agent AI systems, and RAG-based LLM pipelines to real-world problems in the energy market.

My research addresses the growing need for intelligent, secure, and scalable AI systems in modern IT infrastructure and online services, conducted under a Dottorato in Apprendistato (industrial doctorate) in collaboration with Aruba Group. Specifically, I focus on the post-training and fine-tuning of large language models for domain-specific applications, investigating how general-purpose models can be effectively specialized to meet the accuracy and relevance demands of real-world business contexts. A core part of my work involves the development of agentic AI systems capable of autonomously monitoring, reasoning, and acting within complex IT environments, as well as the broader integration of generative AI into diverse industrial workflows. I am also evaluating the trade-offs between open-source and closed-source AI architectures, with the goal of developing secure, modular, and scalable AI frameworks suited to organizations of varying sizes.

Ricerca

Gruppi di ricerca