Federal Project Grant Award Summary

New York University received a $400,000 Project Grant from the National Science Foundation's Division of Computing and Communication Foundations under the Computer and Information Science and Engineering program (CFDA 47.070), awarded on July 15, 2026, with an ultimate completion date of June 30, 2030. This collaborative research initiative focuses on developing improved algorithms for graph-based vector databases to enhance their speed, reliability, and computational efficiency. Vector databases store and enable efficient search and filtering over vector embeddings—numerical representations generated by machine learning systems to encode information about documents, images, videos, and other media. By optimizing the algorithmic foundations of these systems, the research aims to reduce computational costs and resource requirements, thereby democratizing access to vector database technology for small businesses, scientists, and artificial intelligence practitioners with limited cloud computing infrastructure budgets.

The project addresses a critical gap between advances in machine learning and the engineering systems required to operationalize vector embeddings at scale. The research will generate algorithms and design principles applicable to next-generation vector databases capable of handling large-scale data while maintaining performance and cost-effectiveness across diverse applications in web and product search, scientific data management, generative artificial intelligence, and related domains. This work will contribute foundational knowledge to the broader Computer and Information Science and Engineering ecosystem, supporting both research innovation and practical deployment of artificial intelligence technologies across sectors.