Anderson Augusma

I develop deep learning methods for understanding visual and audiovisual data in real-world environments. My work combines computer vision, multimodal learning, and human behavior analysis, with particular attention to privacy-aware perception.

I hold a PhD from Université Grenoble Alpes, where my research focused on group-level emotion recognition in unconstrained videos. My experience includes designing neural architectures, building PyTorch training and inference pipelines, combining video and audio representations, and evaluating models on real-world datasets.

I am interested in research and engineering challenges that connect strong machine learning methods with practical applications. Building on my background in multimodal perception, I am particularly motivated to contribute to robotics, robot learning, and human-robot interaction.

I am seeking research scientist, research engineer, and machine learning engineer opportunities involving computer vision, multimodal AI, or perception. I also welcome postdoctoral opportunities in these areas.

Outside research, I enjoy hiking, going to the gym, and playing music. I previously played trumpet with the Orchestre du Campus de Grenoble.