Teradata unveils tools for building AI agents

Teradata has unveiled a set of tools for enabling enterprises to build and manage autonomous AI agents. AgentBuilder leverages open-source frameworks and the Teradata AI and knowledge platform to allow IT teams to design, operationalise and manage multi-agent systems.

The recently revealed capabilities also include a range of ready-to-deploy Teradata agents based on pre-built templates for tasks such as churn analysis and system monitoring. The pre-built agents will be designed to address domain-specific challenges.

Examples of pre-built agents include the Teradata SQL agent, designed to convert questions or requests expressed in natural language into SQL queries, as well as the Teradata data science agent, which will aim to create executable machine learning pipelines from natural language requests.

AgentBuilder will be available in private preview later this year. The initial release will support open-source frameworks such as Flowise and CrewAI, with LangChain and LangGraph planned to follow.

Teradata chief product officer Sumeet Arora said the solution is aimed at helping customers overcome the barriers to operationalising autonomous agents, including fragmented and siloed data that may lead to hallucinations and unreliable outputs, as well as performance bottlenecks and limitations with governance.

“AgentBuilder represents meaningful progress in advancing agentic AI for the autonomous enterprise,” he said. “By combining the flexibility of open-source frameworks with Teradata’s AI and knowledge platform and our MCP Server, which provides deep semantic access to enterprise data, we’re helping organisations build intelligent agents that are not only autonomous and scalable, but also deeply aligned with their business goals, governance standards and domain expertise.”

Image credit: iStock.com/sankai