Teradata Introduces an Enterprise-Wide Vector Database to Empower Customers in Preparing for the Deployment of Reliable, Autonomous AI Techniques
Teradata, a leading data analytics company, has introduced Teradata Enterprise Vector Store (EVS), an innovative in-database solution designed to manage vector data and support Trusted Artificial Intelligence (AI) applications. This new offering is expected to help enterprises maximize value and insight from unstructured data while reducing spend.
Teradata Enterprise Vector Store is set to revolutionize customer service with its augmented call centre use case. By leveraging agentic AI and Retrieval-Augmented Generation (RAG), EVS enables upsell and cross-sell opportunities during customer interactions, transforming the way businesses engage with their customers.
The key benefits of Teradata Enterprise Vector Store lie in its ability to unify diverse data types for AI applications, accelerate vector computations through GPU integration, and provide a robust, compliant environment for Trusted AI development at enterprise scale. This positions Teradata strongly for organizations leveraging advanced AI and generative models with stringent data governance needs.
One of the standout features of EVS is its integration with NVIDIA NeMo Retriever, focusing on vector data management and Trusted AI applications. This integration provides a powerful vector search and retrieval capability, enabling enterprises to efficiently manage and query vector embeddings derived from AI models. It also allows real-time, context-aware search and analytics across complex data landscapes.
EVS supports frameworks like LangChain and RAG, along with comprehensive data management and governance practices for Trusted AI. It engages with the full lifecycle of vector data management, including metadata management and intelligent search. NVIDIA NeMo Retriever extraction is designed to enable the use of information and insights from unstructured data sources such as PDFs.
Louis Landry, the CTO of Teradata, stated that Teradata Enterprise Vector Store delivers a dynamic, trusted foundation large organizations need for agentic AI. The offering aims to create a single, trusted repository for all data and builds on Teradata's support for retrieval-augmented generation (RAG).
Teradata Enterprise Vector Store is a scalable, in-database vector solution built with NVIDIA AIT. It processes work within the existing Teradata system, which supports flexible deployment options including cloud, on-premises, or hybrid. The solution is designed to cost-effectively deliver the sophistication required for solving complex business challenges and is expected to integrate NVIDIA NeMo Retriever to provide an information retrieval solution with high accuracy and data privacy.
Currently, Teradata Enterprise Vector Store is available in private preview, with general availability expected in July. This new offering from Teradata is set to redefine the way enterprises manage and leverage their data for AI applications, offering a performant way to enable use cases that require vector capabilities and RAG applications.
The Teradata Enterprise Vector Store, integrated with NVIDIA NeMo Retriever, offers a powerful vector search and retrieval capability for diverse data types, particularly beneficial for Trusted Artificial Intelligence applications. This innovation in technology is expected to revolutionize AI-driven customer interactions by leveraging agentic AI and Retrieval-Augmented Generation (RAG), enabling upsell and cross-sell opportunities during customer interactions.
