Harshdip Saha

I'm a pre-final-year Computer Science student, specialising in AI, at Netaji Subhas University of Technology in Delhi, and an undergraduate researcher at NexGenLab NSUT. My work sits where machine learning meets medicine: computer vision, brain imaging, and the pipelines that make a model's answer trustworthy enough to check against a radiologist's.

Most of it started with one question — can a model tell, from two MRI scans months apart, whether a brain tumour is growing? That became RECAP-Net. It finished 3rd worldwide in the BraTS Lighthouse 2025 Tumor Progression Challenge — an international competition where every team's model is scored on the same hidden set of brain-tumour scans — and I presented it as a talk at MICCAI 2025 in South Korea, the main conference for AI in medical imaging.

Since then the work has widened. At IIT Madras I develop PyAMorph, a library that turns images and CAD models into geometry a physics simulation can run on. At Optum (UnitedHealth Group) I spent a summer on the team piloting AI-DLC — a way of building software where every change carries a written plan and record — across a 35,000-person organisation. And in open source I'm one of the top contributors to BrainGlobe, the toolkit neuroscience labs use to align whole-brain microscopy images to a standard atlas.

Away from the keyboard: 1000+ solved programming problems across LeetCode and GeeksforGeeks — mostly dynamic programming and graph problems, which is probably why the segmentation code looked the way it did.