Teradata AgentBuilder FAQ

Teradata AgentBuilder is a set of capabilities that enables enterprises to build and manage autonomous AI agents with contextual knowledge, domain expertise, and hybrid deployment flexibility.

• Analyst validation: Recognized by Gartner and Forrester as top growth area.
• Market demand: Enterprises struggle to get value from AI agents.
• Strategic alignment: It’s a catalyst for Teradata’s future.

Despite growing interest in agentic AI, enterprise decision-makers, data scientists, AI practitioners, and IT leaders continue to face significant barriers to operationalizing autonomous agents. These include fragmented and siloed data that lead to hallucinations and unreliable outputs, a lack of embedded business knowledge that limits contextual relevance, performance bottlenecks and inflated costs from prompt-driven workloads, and insufficient governance that can make deployment risky.

We enable expert agents enriched with industry context and enterprise-wide intelligence to drive more autonomous, informed decision-making.

• Build and Manage Autonomous AI Agents: AgentBuilder enables enterprises to create and orchestrate autonomous AI agents that leverage trusted data, deep domain expertise, and hybrid deployment flexibility (cloud and on-premises).
• Low/No-Code Agent Creation: Provides a user-friendly interface (UI) for building agents and developing multi-agent workflows. Supports open-source tools like Flowise, CrewAI, and LangGraph, as well as integration with partner and cloud service provider tools.
• Pre-Built Expert Agents: Includes a suite of pre-built agents for rapid adoption and immediate business impact, such as:
  • SQL Agent: Converts natural language into optimized SQL queries for Teradata data warehouses.
  • Data Science Agent: Builds fully executable machine learning pipelines from natural language requests.
  • Monitoring Agent: Continuously monitors and manages Teradata databases and infrastructure for health and performance.
• Custom Agent Development: Users can build their own agents and use cases, leveraging Teradata’s Model Context Protocol (MCP) Server for secure, scalable agent deployment. The MCP Server provides connectivity, authorization, and extensibility for additional tools and agent studios (e.g., Bedrock, Azure, GCP).
• Enterprise-Grade Governance: Built-in features for compliance, auditability, and policy enforcement, ensuring responsible and secure deployment of autonomous agents.
• Embedded Business Knowledge: Agents can be enriched with industry context, enterprise-wide intelligence, and reusable assets (like industry data models), enabling more relevant, domain-aware, and business-aligned recommendations.
• Hybrid and Scalable Deployment: Supports deployment across cloud and on-premises environments, with tools for efficient, autonomous data processing and decision-making at scale.
• Open and Extensible Ecosystem: AgentBuilder is compatible with open-source frameworks and allows users to contribute tools back to the product team, fostering continuous improvement and innovation.

Yes, we are providing the following expert agents:

• Customer Lifetime Value agent is an expert agent purpose-built for CLV use cases, empowering data exploration, actionable insights, and strategic decision-making.
• Teradata Data science agent is an expert agent that creates fully executable ML pipelines from natural language requests.
• Teradata SQL generation agent is an advanced expert agent that converts questions or requests expressed in natural language into SQL queries that can be executed against Teradata data warehouse tables.
• System health monitoring agent automates overseeing of Teradata databases, servers and sub-systems to maintain reliability, detect issues and optimize performance.

Customers can leverage an AgentBuilder to autonomously solve business problems. These agents will operate within the structure of the industry framework such as Customer Intelligence Framework and be tightly coupled with our industry data models.

• Trusted, Contextual data at scale: Teradata Vantage unifies structured and unstructured data across cloud and on-prem environments, ensuring AI agents make accurate, real-time, context-aware decisions.
• Embedded business knowledge and deep domain expertise: Agents can use Teradata’s industry expertise and reusable assets to create prompts and workflows that deliver domain-aware, business-aligned recommendations.
• Scalable, efficient tools for autonomous agents in hybrid environment: With 150+ in-engine functions, BYOM/BYO-LLM support, and RAG capabilities, Teradata enables agents to process and act autonomously on data directly within the platform—minimizing latency and cost.
• Enterprise-grade governance: Built-in features for compliance, auditability, and policy enforcement, allowing to deploy autonomous agents responsibly and securely.
• Pre-built agents and expert agents: Pre-built Agents offer ready-to-use templates for tasks like churn prediction and system monitoring designed to help teams get started quickly with agentic AI.

Currently, the AgentBuilder installation is separate from Vantage.    

AgentBuilder requires two installations: a front-end (currently Flowise) and Teradata MCP server, which can connect to VantageCloud Enterprise or VantageCloud Lake.

There is no specific restriction related to version, but it is recommended to be 17.20 onwards to leverage MCP tools related to advanced functionality such as vector store.

Currently, both Flowise and the Teradata MCP server are open-source, so there’s no associated cost. However, new features coming in 2026 will include additional costs.

Customers can join the waitlist for private preview. In the meantime, customers can contact the accounts team for a demo.