About Me

I work as a compute architect/researcher at Nvidia. I dabble with topics related to precision, sparsity, training efficiency and inference deployments of small and large DL networks.

Till recently, I worked as an AI/ML Engineer at Apple. At Apple, I worked on improving the performance of AI/ML applications on Apple's Si devices. I also worked on optimizing neural network models powering iOS Camera in iPhone 14 and iPhone 15 devices. 

Previously, I worked as a DL Computer Architect at NVIDIA. In this role, I worked on sparsity in DL networks and Sparse Tensor Cores. More details can be found in this article.

Prior to this, I worked as a Research Scientist in Intel Labs' Accelerator Architecture Lab.  At Intel I worked on improving Intel's Xeon and Xeon Phi's performance for machine learning workloads (new CPU AVX-512 instructions for Xeon and Xeon Phi cores). 

Prior to this, I was a PhD student in the Department of Computer Science and Engineering at Penn State University. For my PhD thesis, I worked in the area of Network-on-Chip (NoC) Architectures. My thesis advisor is Chita Das. I also collaborated with professors Vijay Narayanan (Penn State), Mahmut Kandemir (Penn State), Yuan Xie (Penn State) and Onur Mutlu (ETH). 

My research interests lie mostly in the area of AI/ML/DL applications and computer micro-architecture. I am also interested in cache architectures for Chip Multi-Processors (CMPs), 3D architectures, resource provisioning and power management aspects in CMPs, and Internet-Scale Data Centers.

I hail from a tiny city called Rourkela in India. Prior to joining Penn State, I completed my undergraduate studies from National Institute of Technology, Rourkela in India. While in grad school, I was a photographer for the local, student run newspaper - Daily Collegian. I am passionate about sports and portrait photography.