Teradata Analyst Agent

The Teradata Analyst Agent provides AI-assisted decision support, enabling business users to explore enterprise data and generate insights using natural language.

Users can ask questions and receive analysis, visualizations, and key insights generated directly from their data, without needing to write SQL.

The agent connects securely to Teradata Vantage using Teradata Enterprise Model Context Protocol (MCP), allowing organizations to analyze enterprise-scale data without moving or duplicating it. Built-in Agent Telemetry provides transparent, governed AI operations by capturing execution, performance, and token usage data for every request.

This enables organizations to continuously monitor and optimize the agent while supporting responsible AI and governance as AI-assisted analytics expands.

Key Features

• Interactive Frontend – Supports multiple concurrent chat sessions, chat history, export of charts, and user feedback capture for each response.
• Intelligent Visualization Engine – Generates visualizations to support insights, by selecting from 12+ chart types.
• Autonomous Backend – Multi-agent orchestration coordinating schema discovery, SQL generation, query execution, statistical analysis, and visualization through MCP-based tool integrations.
• Agent Telemetry – Captures request-level execution data including agent calls, token usage, response times, tool activity, cost metrics, quality signals, and guardrail events.
• Configurability – Supports integration with native Azure services, swappable LLM providers, configurable guardrails, and environment-level configuration for models and agent components.

Deployment:

• The Teradata AI Services team works directly with the customer to finalize cloud architecture, security configuration, and storage requirements before executing an automated deployment into the customer's Azure subscription.