About me

Hello! I’m Gabriel Adriano de Melo, an aspiring AI researcher and computer engineer with a passion for pushing the boundaries of artificial intelligence. I’ve worked on a wide range of projects, from developing computer vision systems for aeronautical applications to exploring some theoretical aspects of large language models. Each experience has fueled my curiosity and deepened my commitment to advancing the field.

Currently, as a PhD Student at ITA (Instituto Tecnológico de Aeronáutica), where I also received his MSc in 2020 in the same area and BSc in Computer Engineering in 2019, I’m focusing on one of the most critical challenges in AI: ensuring the safety and reliability of large language models. I’m particularly intrigued by innovative approaches to improve model robustness and generalization. Some of the ideas I’m exploring include dynamic network expansion during training, sparsity techniques, and the implementation of internal loops in models.

My research isn’t just about technical advancements, though. At its core, it’s driven by a desire to create AI systems that are not only powerful but also align with human values and intentions. I believe that as we push the capabilities of AI, we must also ensure that these systems are safe, reliable, and beneficial to society.

Throughout my academic and professional journey, I’ve had the opportunity to wear many hats - from research engineer at the Flight Test and Research Institute to co-lecturer for a computer vision course at ITA. Each role has broadened my perspective and reinforced my belief in the importance of collaboration and knowledge sharing in driving innovation.

When I’m not deep in research or coding, you might find me pondering the philosophical implications of AI or engaging in discussions about the future of technology. I’m always eager to connect with fellow AI enthusiasts, whether it’s to collaborate on exciting projects or simply to exchange ideas about the latest developments in the field.