Vinu Joseph is a senior research scientist on the Programming Systems and Applications research team at NVIDIA. He completed his Ph.D. in computer science at the University of Utah where his research was focused on resource-efficient Neural Network Compression with Correctness Emphasis. With a graduate research fellowship from NVIDIA, he worked on Condensa: A Programming System for Neural Network Compression. Before graduate school, he was an engineer at Arm Inc. His research goal is to develop tools and techniques that accelerate encrypted computing, fully homomorphic encryption, and post-quantum crypto algorithms so that they can be used to develop resource-efficient and privacy-preserving generative AI frameworks.