Microsoft: Data Formulator 0.7 connects fragmented enterprise data

Microsoft Research has released Data Formulator 0.7, an open-source AI system for enterprise data analytics. The system addresses fragmented workflows by connecting different data sources, agents and visualization in a single workspace.

What does Data Formulator solve?

The system connects fragmented enterprise data and iterative analytical workflows. Fragmented data is data scattered across various systems — databases, warehouses and files — that otherwise has to be manually joined for every analysis. Data Formulator unites this into a single working environment.

What are persistent data connectors?

Persistent connectors establish lasting links between an enterprise’s data sources and Data Formulator — across databases, warehouses, BI systems, object stores and local files. This removes repetitive manual work because the links are reusable and centrally managed.

How do the AI agents work?

Unlike isolated chatbots, the context-aware agents access the entire workspace — connected data, loaded tables, previous visualizations and user goals. They can browse data, write and execute code in an isolated environment, and generate chart specifications with verifiable, reproducible code. The interface combines the Data Thread (a structured chat that preserves the analysis history) and the Interactive Canvas for editing visualizations with natural language. The key advantage is that the work requires no knowledge of SQL or programming.