AI Researcher · Human
Embodied & Physical AI · Video Understanding · Spatial Reasoning
Hi, I am Shivam Singh, an AI researcher with 3+ years of research experience across IIIT Hyderabad and AMD. I completed my Masters at Robotics Research Centre, IIIT Hyderabad, where I worked on embodied AI, classical planning, task anticipation, and human-robot collaboration. I later joined AMD's AI Models & Agents team, where my work shifted toward Vision-Language Models, video understanding, multimodal reasoning, and memory. These days, I am particularly interested in how intelligent systems can perceive, reason about, and act in the world over long horizons. My recent work focuses on video understanding and spatial reasoning — the idea that a machine could watch ordinary video and build a useful representation of space, memory, and meaning that allows it to reason about its environment.
I have always loved taking pictures and making videos. At the same time, I am fascinated by how the human mind works — how we build abstract representations of incredibly complex things and use them to reason about the world. That is what motivates me most in this field: the intersection of my love for capturing the world and my curiosity about how intelligent systems turn messy, complex reality into something they can understand and reason about. I think the most interesting problems sit at the intersection of perception, reasoning, memory, and action, and I enjoy working on them with curious, generous people.
I am always open to collaboration. If any of this resonates with you — or you simply want to talk — just ping me at shivamsingh3458@gmail.com and we can take it from there.