Karthik Garimella, a Ph.D. student in the NYU Tandon Department of Electrical and Computer Engineering, was recently named a “Rising Star” by MLCommons, an AI consortium dedicated to “encouraging open collaboration and improving the accuracy, safety, speed and efficiency of AI technologies.” The “Rising Star” designation honors young researchers working at the intersection of machine learning and systems, and supports their development through access to workshops, networking opportunities, and other resources. According to Vijay Janapa Reddi, MLCommons VP and Research Chair and steering committee member of the Rising Stars program, “by nurturing and supporting the next generation of researchers, both domestically and globally, we aim to foster an inclusive environment where these individuals can make groundbreaking contributions that will shape the future of ML and systems research.”
Garimella, who came to Brooklyn for his doctoral studies after earning a bachelor’s degree in physics from Hendrix College and a master’s degree in computer engineering from Washington University in St. Louis, earned the honor for his investigative work in protecting the privacy and security of machine-learning technologies. Now working under the guidance of Assistant Professor Brandon Reagen, a faculty member of NYU’s Center for Cybersecurity, Garimella is particularly focused on closing a “privacy loophole” created in the use of cloud-based applications, including social media. As he explained to NYU News, this loophole is created by the way AI shares information. “You type in a prompt, and it gets sent to the cloud, and although it’s secured during transmission, at the end of the day, the company still has access to it. This is concerning, especially if it contains sensitive or personal information.” To address this loophole, Garimella is using a number of technologies, including multi-party computation, where differing parties can pool data for computing tasks and access the resulting output without revealing information from any individual party, and homomorphic encryption, a technique that enables mathematical operations to be performed on encrypted data without affecting the encryption.
His advisor was unsurprised that Garimella, who is currently interning at NVIDIA, was chosen for the competitive program.”He is a fantastic student whose academic talent, rigor, and creativity exemplify everything we hope for in a Tandon doctoral candidate,” Dr. Reagen comments. “ I predict that his work will make a major impact.”
In addition to his “Rising Star” designation, Garimella was also the recipient of a 2024 Li Publication Award, given by the NYU Tandon Department of Electrical and Computing Engineering.