Nitthilan Kannappan Jayakodi

Papers

Total Citations

H-Index

About

Nitthilan Kannappan Jayakodi is a leading researcher at the forefront of energy-efficient edge AI, specializing in the co-design of hardware and software to bring deep neural networks (DNNs) to mobile and resource-constrained platforms. His work directly addresses the critical challenge of deploying complex AI—from object detection and robotics to augmented reality—on devices where battery life and computational power are limited. Jayakodi’s major contributions include pioneering novel frameworks for runtime energy-accuracy tradeoffs, as demonstrated in his highly cited 2020 work (31 citations) on optimizing pretrained deep models for mobile platforms. He further advanced the field with a general hardware-software co-design methodology (12 citations) for edge AI, enabling efficient inference for applications like self-driving cars and mobile health. His innovative research also extends to 3D vision, with his work on PETNet exploring the tradeoff between polygon count and energy for producing 3D objects from images, a key enabler for AR and VR on mobile devices. Through his focused and impactful body of work, Jayakodi is shaping a future where intelligent, real-time AI is not confined to the cloud, but operates seamlessly and sustainably on the devices in our pockets.