Yash Khandelwal

I'm a Research Engineer in Computer Vision and Machine Learning, currently working on computer vision applications in healthcare at Wadhwani AI. My work focuses on developing multi-task deep learning models for anthropometric measurements of newborns using smartphone videos, aiming to address malnutrition monitoring challenges in developing countries. Previously, I worked on various computer vision and robotics projects at IIT Kanpur, where I helped develop autonomous navigation systems and real-time perception algorithms for intelligent ground vehicles. I've collaborated with researchers across institutions to work on problems spanning crowd counting from drone imagery, visual question answering, and autonomous driving safety systems. My research interests lie at the intersection of computer vision, deep learning, and their real-world applications, particularly in healthcare and robotics. I'm passionate about developing efficient and deployable AI solutions that can create meaningful social impact. My recent work includes NurtureNet, a lightweight multi-task model for contactless newborn anthropometry that can run on low-cost smartphones, achieving clinically viable accuracy while maintaining real-time performance.