🌎 Based in Campinas, SP, Brazil 🎓 PhD Student in Computer Science @ Unicamp 🔧 Software Engineer | 🤖 Researcher in Machine Learning & Computer Vision 🌐 Fluent in English and Portuguese | 🗾 Studying Japanese (N5 -> N4)

I am a researcher and software engineer with a strong interest in machine learning, computer vision, and weakly supervised learning. My background combines solid software engineering foundations with applied research experience, allowing me to bridge the gap between theoretical advances and practical implementations.

👨‍🎓 I hold a Bachelor's degree in Software Engineering and a Master's degree in Computer Science, where I specialized in deep learning and Transformer-based architectures for seismic facies segmentation.

📚 Currently, I am pursuing a PhD in Computer Science at the University of Campinas (Unicamp). My research focuses on:

• Weakly Supervised Semantic Segmentation (WSSS), with emphasis on inaccurate and noisy label supervision
• Pseudo-label refinement and noise-robust learning strategies
• Deep Active Learning for label efficient training
• Foundation model-assisted segmentation under weak supervision

🛠️ I am also one of the core maintainers and software architect of the open-source project Minerva, which provides scalable and flexible tools for deep learning research. My work on Minerva bridges research and engineering by enabling reproducible experiments, large-scale training, and model debugging.

• Weakly Supervised Semantic Segmentation under noisy and inaccurate supervision
• Pseudo-label generation, correction, and refinement
• Deep Active Learning for efficient annotation and model improvement
• Transformer architectures for computer vision
• Applied AI research at the intersection of software engineering and machine learning

I am open to collaborations and discussions on topics related to machine learning, weakly supervised learning, and applied AI research.

📫 Feel free to reach out via email at gabriel.bgs00@gmail.com or connect with me on LinkedIn.