This inventor holds 1 USPTO granted patent. Top assignee: Nvidia Corporation. Active years: 2026.
Data source: USPTO (United States Patent and Trademark Office) public patent records. Weekly synchronization. How IDiyas builds this profile
Company Filing History:
Years Active: 2026
Title: Innovations by Nikita Shustrov
Introduction
Nikita Shustrov is an accomplished inventor based in Santa Clara, CA. He is known for his contributions to the field of matrix computations and data processing. His innovative work has led to the development of a patent that enhances the efficiency of matrix multiplication operations.
Latest Patents
Nikita holds a patent for "Non-rectangular matrix computations and data pattern processing using tensor cores." This patent describes a method where matrix multiplication operations can be implemented on tensor cores of a parallel processing unit. The efficiency of these operations is improved when one of the input operands or the output operand is a square matrix with a triangular data pattern. By dropping computations and masking out elements of the square matrix input operand on one side of the main diagonal, the number of computations performed can be significantly reduced. Additionally, the patent provides a library that implements these matrix multiplication operations.
Career Highlights
Nikita Shustrov is currently employed at Nvidia Corporation, a leading company in the field of graphics processing and computing technology. His work at Nvidia focuses on enhancing computational efficiency through innovative algorithms and processing techniques.
Collaborations
Nikita collaborates with Aniket Shivam, a talented coworker who contributes to their shared projects and innovations.
Conclusion
Nikita Shustrov's work exemplifies the spirit of innovation in the field of matrix computations. His patent not only advances technology but also showcases the potential for improved efficiency in data processing.
Data source: USPTO (United States Patent and Trademark Office) public patent records. Weekly synchronization. How IDiyas builds this profile
