Anubhab Das

I am a third-year B.Tech. student in Electronics and Instrumentation Engineering at the National Institute of Technology Rourkela. My research interests lie in deep learning, computer vision, and image restoration, with a focus on designing efficient and reproducible neural models.

At NIT Rourkela, I am currently working on underwater image enhancement, exploring deep learning–based restoration techniques for degraded visual data captured in challenging aquatic environments. I am particularly interested in bridging the gap between theoretical advances and deployable, resource-efficient systems for real-world visual understanding.

Beyond research, I enjoy implementing and experimenting with deep learning systems—from building lightweight autograd engines to training and optimizing neural architectures. A central aspect of my development philosophy is computational minimalism—achieving strong performance with lean, well-engineered code. I primarily work on Arch Linux, maintaining a custom, minimal setup using Hyprland, optimized for efficient experimentation and clean research workflows.

In the long term, I aim to pursue research that advances the intersection of efficient visual learning and practical AI deployment, contributing to systems that are both scientifically grounded and computationally accessible.