---
title: "From Data to Dialogue: How S&P Global Energy Made Its Structured Data Estate Conversational with Databricks Genie Agents and MCP"
description: "S&P Global Energy aimed to make its structured data estate—spanning LNG, Chemicals, Crude Oil, and other commodities—conversationally accessible via AI agents. The team evaluated approaches and chose Databricks Genie Agents, where subject matter experts curate one agent per dataset group without writing code. Each Genie Agent is automatically exposed as a managed MCP server, providing a governed, zero-deployment integration point. A FastMCP-based proxy composes multiple Genie MCP servers into composite endpoints for cross-domain questions. This architecture shortened time to market from months to days, with governance via Unity Catalog and an open MCP standard enabling both internal and external consumers."
---

# From Data to Dialogue: How S&P Global Energy Made Its Structured Data Estate Conversational with Databricks Genie Agents and MCP

[Databricks](https://yomu.fyi/company/databricks) · Debaprasad Satapathy, Eric Henderson · Sep 25, 2026

**Type:** Problem & solution

## Summary

S&P Global Energy aimed to make its structured data estate—spanning LNG, Chemicals, Crude Oil, and other commodities—conversationally accessible via AI agents. The team evaluated approaches and chose Databricks Genie Agents, where subject matter experts curate one agent per dataset group without writing code. Each Genie Agent is automatically exposed as a managed MCP server, providing a governed, zero-deployment integration point. A FastMCP-based proxy composes multiple Genie MCP servers into composite endpoints for cross-domain questions. This architecture shortened time to market from months to days, with governance via Unity Catalog and an open MCP standard enabling both internal and external consumers.

## Context

S&P Global Energy’s structured data estate spans multiple commodities (LNG, Chemicals, Crude Oil, etc.) with complex datasets. The goal was to enable natural language access for external customers via AI agents while maintaining governance. Traditional approaches like hand-built text-to-SQL or custom APIs were too slow and brittle.

## Approach / What changed

Using Databricks Genie Agents curated by subject matter experts per dataset group, exposed as managed MCP servers, and composed via a FastMCP proxy for cross-domain queries.

## Takeaways

- SMEs curate one Genie Agent per dataset group without code, adding context like column descriptions and example queries to make text-to-SQL reliable.
- Each Genie Agent is automatically a governed MCP server with two tools (query and poll), requiring no deployment.
- A FastMCP proxy composes multiple Genie MCP servers into composite endpoints, enabling cross-domain questions while maintaining governance.

## FAQ

### How do subject matter experts create Genie Agents without engineering involvement?

SMEs select relevant tables (native or federated), group them by dataset group, and add context like column descriptions and example queries. No code is required.

### How are Genie Agents exposed as MCP servers?

Each Genie Agent is automatically a managed MCP server at a Databricks endpoint, exposing two tools: genie\_query\_space and genie\_poll\_response. No deployment or hosting needed.

### How does the architecture handle cross-domain questions?

A FastMCP-based proxy composes multiple Genie MCP servers into composite endpoints, allowing an agent to query across domains like LNG cargoes and chemicals supply-demand.

**Tags:** [AI Agents](https://yomu.fyi/topic/ai-agents), [Databricks](https://yomu.fyi/topic/databricks), [Data Governance](https://yomu.fyi/topic/data-governance), [MCP](https://yomu.fyi/topic/mcp)

- Source: [Databricks](https://www.databricks.com/blog/data-dialogue-how-sp-global-energy-made-its-structured-data-estate-conversational-databricks)
- Source URL: https://www.databricks.com/blog/data-dialogue-how-sp-global-energy-made-its-structured-data-estate-conversational-databricks
- Ingested by Yomu: 2026-09-25T18:00:08.097Z

[Read original post](https://www.databricks.com/blog/data-dialogue-how-sp-global-energy-made-its-structured-data-estate-conversational-databricks)