#data-agent

Jan 13, 2026

Analyzing Microsoft Fabric's Data Agent

This article analyzes data agents on a variety of tasks and queries and performs some prompt engineering to find out the change in responses.

Apr 27, 2026

Building a Fabric Data Agent with On‑Premises SQL Server Data

Many organizations still rely on on‑premises SQL Server for core operational workloads — finance, billing, ERP, manufacturing, and regulatory systems — that cannot be easily migrated to the cloud. At the same time, the business is demanding: Conversational access to data (“Ask the data” experiences) AI‑powered insights without building fragile ETL pipelines Faster time‑to‑value from analytics and AI Strong governance, security boundaries, and system ownership Microsoft Fabric addresses this challenge by allowing organizations to bring AI to their data before moving their data to AI. By leveraging SQL Server mirroring into Microsoft Fabric, organizations can continuously replicate on-premises data into OneLake, making it immediately available for analytics and AI—without disrupting source systems. On top of this foundation, semantic models and Fabric Data Agents enable governed, natural-language interactions powered by Microsoft Copilot.

Jan 1, 2026

Chat with your data using Microsoft Fabric data agents

Chat with your data using Microsoft Fabric data agents A Microsoft Fabric data agent enables natural interaction with your data by allowing you to ask questions in plain English and receive structured, human-readable responses. By eliminating the need to understand query languages like SQL (Structured Query Language), DAX (Data Analysis Expressions), or KQL (Kusto Query Language), the data agent makes data insights accessible across the organization, regardless of technical skill level.

Mar 14, 2026

Building a Spider2-Inspired Benchmark to Measure the Real Robustness of a Fabric Data Agent in Italian

This article moves from working demos to measurable reliability by introducing a Spider2-inspired benchmark for evaluating a Fabric Data Agent in Italian. It explains why manual spot checks are not enough, and shows how to design a more rigorous evaluation framework that separates already-taught patterns from true generalization. The result is a practical benchmark design for assessing multilingual Fabric Data Agents beyond isolated successful examples.

Jan 9, 2026

Creating Data Agent in Microsoft Fabric — First Impression

To give you a bit of background on what this data agent is all about, think of it as your assistant that knows everything about your data. Just like you use ChatGPT, you can leverage LLM capabilities on your enterprise data and have conversations over your database.

Jan 12, 2026

Creating Data Agent in Microsoft Fabric - First Impression

The article walks us through the creation of data agent for first time user and explores various ways to play around data agent.

Jan 19, 2026

Creating Microsoft Fabric Data Agent using python sdk programmatically

This article explores creation of Data Agent programatically using Python SDK for data agents.

Jan 1, 2026

Data Agent in Microsoft Fabric – Here’s How it Works by Azure Ops

Data Agent in Microsoft Fabric – Here’s How it Works Ever wished you could just ask your data questions in plain English and get instant, intelligent answers? With Microsoft Fabric’s new Data Agent, that’s not just possible, it’s powerful. In this post, I’ll walk you through how I built a Fabric Data Agent on top of the standard AdventureWorksDW dataset, and how you can too, even if you’re a complete beginner.

Apr 1, 2026

Designing AI‑Ready Analytics with Microsoft Fabric Data Agents

s organizations accelerate their adoption of AI-powered analytics, a new challenge is emerging: how to expose trusted business data to AI—Copilot, data agents, and conversational analytics—in a way that is secure, cost-predictable, and architecturally sound. Most enterprises are already operating in a complex reality: Power BI reports connected to operational systems like Oracle, SAP, and SQL Server A mix of Import and DirectQuery semantic models Strong governance expectations around security, lineage, and cost control

Mar 4, 2026

Fabric Data Agents Are English-First (For Now): A Hands-On Guide to Configuring One on Zava DIY for Non-English Users

This article provides a hands-on, incremental guide to configuring a Microsoft Fabric Data Agent on the Zava DIY dataset for non-English users, while keeping the agent grounded in an English-first setup. It shows how to improve reliability step by step through data source descriptions, agent instructions, domain constraints, formatting rules, and validated example queries, then extends the configuration with a practical "translate in, translate out" approach. The result is a reproducible quick-win pattern for making the agent more analytics-ready across languages without introducing external translation layers or custom front ends.

Mar 26, 2026

Fabric Data Agents: The Shift from Querying Data to Reasoning Over Knowledge

What’s the big deal? Fabric is no longer just a data platform. It’s becoming a system that can reason over enterprise knowledge. The real shift is the move from human-authored to AI-authored logic, under guardrails. Before Humans write explicit queries, design semantic models, orchestrate pipelines. Tools act as passive executors, doing exactly what they are told. Now Humans declare the intent. Agents determine the plan (how to answer) and execute across governed data, semantic and orchestration layers.

Mar 31, 2025

Implementing Data Agent in Microsoft Fabric for comprehensive business insights

Leveraging the semantic model for business insights The semantic model in Microsoft Fabric plays a crucial role in transforming raw data into meaningful business insights. By accessing the semantic model, Data Agent can understand and interpret complex data relationships, providing a unified view of the data that aligns perfectly with current reporting and analytical needs.

Aug 12, 2025

Microsoft Fabric Data Agent — Complete Guide by Alpa Buddhabhatti

Unlock natural language access to your enterprise data with Microsoft Fabric Data Agent.

Apr 7, 2026

New article: Which Verdicts Changed, and Why: a Row-Level Audit of Fabric Data Agent Evaluation

The author performs a detailed row‑level audit of a 72‑question benchmark to understand why evaluation verdicts changed after fixing errors in the benchmark itself. Many initial “failures” turn out to be caused by faulty ground truth, ambiguous phrasing, or inconsistent casing rules rather than true Data Agent mistakes. After refining benchmark wording, tightening Agent instructions, and clarifying metric definitions, accuracy rises to 97.2%. The few remaining errors stem from extremely complex multi‑step prompts and ambiguous schema references, revealing limits of the underlying model rather than flaws in the benchmark.

Apr 2, 2026

Programmatically Retrieve Prep Data For AI Configuration of Semantic Models

For Power BI Copilot and Data agents with semantic models, you must use Prep Data for AI configuration to ground the responses in the context added in Prep for AI. In this blog, I will show you how you can use the Power BI remote MCP server to get the configuration.

Apr 25, 2026

The Agents: Data Agent & Operations Agent in Practice

A solutions architect's deep dive into the two agents powering Fabric IQ - Data Agents and Operations Agents. Covers the execution chain, governance model, billing math, configuration levers, and the combined architecture where one agent detects and the other explains over a shared Ontology.

Jan 2, 2026

Using Microsoft Fabric Data Agent in Non-English Languages: A Practical Exploration

This article examines what Microsoft Fabric Data Agent's current non-English limitation means in practice, using Italian as a concrete business scenario. Rather than stopping at the official "English-first" guidance, it presents three pragmatic patterns for enabling multilingual experiences today: English instructions with translate-in/translate-out behavior, Copilot Studio as a multilingual front-end, and a translation gateway built around the Data Agent API. The goal is to help teams choose the right architecture for multilingual adoption without overestimating native language support.

Mar 17, 2026

We Built the Benchmark. Now Let’s Evaluate the Fabric Data Agent for Real

This article shows how to move from a benchmark design to a real evaluation workflow for a Microsoft Fabric Data Agent. Starting from a 72-question benchmark built in a previous article for an Italian multilingual scenario, it explains how to complete the ground-truth dataset, run evaluate_data_agent on Fabric, inspect summary and row-level results, and use notebooks to operationalize the full process. A key insight is that part of the observed weakness may come not only from the Data Agent, but also from the evaluation layer itself. By inspecting the SDK source code and testing a stricter custom critic prompt, the article shows how evaluation reliability can improve significantly without changing the agent or the benchmark. Overall, the piece is a practical guide to benchmarking and evaluating Fabric Data Agents more rigorously, especially in multilingual business scenarios.

Mar 26, 2025

What We Learned Building a Real-World Fabric Data Agent

The post is a detailed, field‑tested account of building a real‑world Fabric Data Agent on top of a complex industrial supply‑chain dataset, highlighting what works, what breaks, and what requires deliberate engineering. It walks through six major lessons: why agent instructions—not data sources—determine answer quality; how Semantic Models and Ontologies complement each other; the pitfalls of asynchronous graph rebuilds; the importance of using binding property names in GQL; common GQL query‑generation failures; and the hidden requirement to use dbo. prefixes in SQL. Despite the rough edges of a preview product, the author argues that Fabric IQ’s architecture—unifying SQL, DAX, and GQL behind a governed semantic layer—is powerful and worth understanding deeply for anyone building enterprise‑grade AI analytics.

Data Agent Python SDK Notebooks

Notebooks to evaluate, automate, test data agents

Data Agent Workshop : FabCon EU 2025

Hands-on workshop material from FabCon Vienna 2025 on building data agents

Fabric Data Agent Demo — Wide World Importers

Two sample Microsoft Fabric data agents built on the WideWorldImportersDW data warehouse. A simple agent with minimal instructions silently returns wrong answers for complex queries, while an advanced agent with full data model documentation handles them correctly — same LLM, same data, different instructions. The demo proves that agent instruction quality is the single biggest lever for accuracy when building data agents over enterprise data warehouses with SCD2 dimensions, bridge tables, and weighted allocations.

Ontology with Data Agent

This project provides a complete, end‑to‑end walkthrough for building a Microsoft Fabric Ontology demo—either by installing a ready‑made healthcare example or by creating your own custom ontology, lakehouse tables, demo data, and graph queries. It shows how to use tools like VS Code, GitHub Copilot, Semantic Link Labs notebooks, and the Fabric Ontology Playground to generate RDF models, orchestrate table creation, build a graph database, and construct multi‑hop GQL queries that highlight the strengths of graph reasoning. It also guides you through creating a Fabric Data Agent powered by your ontology, including AI instructions and sample prompts, enabling a fully functional demo environment in under an hour.

Apr 18, 2026

Best Practices for Data Agents in Microsoft Fabric

A deep dive into building and leveraging Data Agents within Microsoft Fabric — AI-powered autonomous systems that integrate with Fabric's lakehouse, warehouses, and semantic models to intelligently query, transform, and analyze data using natural language. This session will explore how Fabric's agent capabilities streamline data workflows and how to configure data agents with the right context to improve accuracy and reliability.

Apr 16, 2026

Fabric Data Agents Series (1/4): From Queries to AI Reasoning in Microsoft Fabric

This session kicks off our 4-part series on Microsoft Fabric Data Agents, exploring a major shift in how we interact with data — from writing queries to AI-driven reasoning under governance. In Part 1, we break down what Fabric Data Agents are (and are not), how they fit into the Microsoft Fabric ecosystem, and how the agent reasoning loop plans and executes grounded, trustworthy answers across governed data. You’ll learn why this shift matters, how it impacts Power BI developers and data teams, and what it takes to move from prompt-level experimentation to system-level AI thinking. 🚀 This is your starting point to understanding how semantic models, architecture, and governance come together to enable reliable, explainable, and scalable AI-powered analytics. 👉 Stay tuned for Parts 2–4, where we dive deeper into design patterns, architecture, and operating Data Agents at scale. 👍 Like, comment, and subscribe for more deep dives on Microsoft Fabric, Power BI, and AI in data. #MicrosoftFabric #PowerBI #FabricDataAgents #AIinAnalytics #DataPlatform

Apr 26, 2026

Fabric Data Agents Series (2/4): Why Your Data Model Breaks AI in Microsoft Fabric

👉 If your data model isn’t solid, your AI won’t be either. In Part 2 of our Microsoft Fabric Data Agents series, we focus on the foundation that makes or breaks AI: your semantic model and data design. This session shows how to build AI-ready data foundations so Fabric Data Agents can deliver accurate, consistent, and trustworthy results — not hallucinations or ambiguity. 🎯 What you’ll learn: ✔️ Why naming conventions and clear definitions are critical for AI reasoning ✔️ How explicit measures improve consistency and trust ✔️ The role of star schema design in scalable AI analytics ✔️ Common failure modes that break Data Agents ✔️ Practical Fabric-native fixes to improve grounding quality Most AI projects don’t fail because of AI — they fail because of bad data modeling. This session gives you the frameworks to fix that. 🚀 Whether you're a Power BI developer, data engineer, or architect, this is your blueprint for moving from experiments → production-ready AI systems. 👉 Watch Part 1 to understand Fabric Data Agents fundamentals 👉 Continue with Parts 3 & 4 to master architecture and governance at scale 👍 Subscribe for more deep dives on Microsoft Fabric, Power BI, and AI-driven data platforms #MicrosoftFabric #PowerBI #DataModeling #FabricDataAgents #AIinAnalytics

Jun 4, 2025

How to Set Up Fabric Data Agents in Microsoft Fabric

Discover how to configure Fabric Data Agents to streamline and manage your data movement across environments in Microsoft Fabric. In this tutorial, we’ll guide you through the installation, setup, and key use cases for Data Agents, helping you connect to on-premises sources and automate secure data flows. Perfect for Fabric admins and data engineers looking to enable hybrid data scenarios with ease. Hi, I’m David, the Managing Director of Level Up Your Data based in Brisbane, Australia and a Microsoft Data Platform MVP. I am passionate about the positive impact data can have on our lives, businesses, world economies, and the environment and enjoy sharing my knowledge with you.

May 26, 2026

How to use Data Agent Example Queries

Example Queries are THE key value differentiator between Microsoft Fabric Data Agents and all other RAG agents you can build, so let's dive into this topic with Bradley Ball, aka ‪@SQLBalls‬ , and look and how we can use them!

Oct 15, 2025

Microsoft Fabric の AI 機能デモ

このビデオは、Fabric の ミラーリング、ショートカット、AI 関数、データエージェントやCopilotに焦点をあてています。 データの統合にはじまり、AIを使用した顧客フィードバックの感情分析や、AIエージェントによるデータ分析をFabric上で行う方法を紹介します。 English :This video focuses on Fabric features such as Mirroring, Shortcuts, AI functions, Data Agents, and Copilot. It introduces how to start with data integration, use AI for sentiment analysis of customer feedback, and perform data analysis on Fabric with AI agents.

Apr 23, 2026

Microsoft Fabric: How to use Data Agent Example Queries

Example Queries are THE key value differentiator between Microsoft Fabric Data Agents and all other RAG agents you can build, so let's dive into this topic with Bradley Ball, aka ‪@SQLBalls‬ , and look and how we can use them!

Mar 31, 2026

Microsoft Fabricで考える非構造データのAI活用

このビデオは、G-Corporationの永田凌真氏によって紹介され、Microsoft Fabricエコシステム内で非構造データを活用するAIの活用に焦点を当てています。セッションでは、AI FunctionsやData Agentを含む最新のデータアーキテクチャとAI実装戦略について、技術的な詳細を深く掘り下げています English : This video, presented by Ryomaru Nagata from G-Corporation, focuses on leveraging AI for unstructured data within the Microsoft Fabric ecosystem. The session provides a technical deep dive into modern data architecture and AI implementation strategies, including AI Functions and Data Agent

Apr 23, 2026

Designing AI‑Ready Data Surfaces: Why Agents Expose Your Weak Spots

Part 2 or 4: Build the semantic and data foundations that make agents reliable. This workshop translates modeling best practices into agent-ready guidance: naming and definitions, explicit measures, star schema discipline, and how to prevent ambiguous data surfaces. Includes common failure modes and Fabric-native remediation strategies to improve grounding quality.

Apr 16, 2026

Designing Trusted Fabric Data Agents - Part 1

This four‑part series guides attendees through the intricacies of Microsoft Fabric Data Agents - from understanding how agents reason over data, to designing AI‑ready data foundations, selecting the right consumption architectures, and operating agents safely at scale. Participants will learn how to move beyond prompt‑level experimentation to system‑level thinking, how agent reasoning, semantic modeling, architectural patterns, and governance controls work together to produce reliable, explainable, and compliant AI‑driven outcomes. Each session builds on the last, equipping teams with practical frameworks, patterns, and checklists to confidently deploy Fabric Data Agents across diverse audiences while managing risk, trust, and performance. Part 1 or 4: A guided tour of why Fabric Data Agents represent a shift from human-authored queries to AI-authored execution logic under guardrails. Learn what a data agent is (and is not), how it fits into the Fabric ecosystem, and how the agent reasoning loop plans and executes grounded answers across governed Fabric artifacts.

May 20, 2026

Fabric Data Agent in a Day

Fabric Data Agent in a Day is a hands-on half-day workshop on Microsoft Fabric Data Agent, scheduled for 20 May 2026 in Milan, designed to show how to move from raw data ingestion to conversational agents that can answer business questions in natural language. During the session, participants populate a Lakehouse and a SQL Database in Fabric, build a first SQL-based Data Agent, make it more effective for Italian-language queries, apply Row Level Security, and measure its performance with Microsoft’s evaluation tools. The workshop then moves to a second agent built on a semantic model with DAX, so attendees can compare the semantic-model approach with the SQL-based one. Overall, the workshop is meant for data and BI professionals who want a practical introduction to building secure, multilingual, end-to-end conversational AI experiences on top of Microsoft Fabric data, using patterns that are closer to real projects than to simple demos.

May 7, 2026

Part 3 / 4: Fabric Data Agent Architecture Patterns: Choosing the Right Consumption Path

Part 3 or 4: Explore the most common and successful architectural patterns for consuming Fabric Data Agents: native Copilot experiences, custom business Q&A apps, agent-to-agent developer workflows via MCP, and enterprise data APIs with optional agent translation. Learn how to choose between Data Agent API, MCP server, and GraphQL depending on your consumers and risk profile.

Data Agent MCP

The Model Context Protocol (MCP) server is an emerging standard in the AI landscape that allows AI systems to discover and interact with external tools in a structured way. It plays a critical role in enabling AI models to access and use external knowledge and capabilities. By using MCP servers, AI systems can extend beyond their own data and reasoning. MCP servers provide a way to expose tools and services to AI systems in a consistent, discoverable manner. They help organizations integrate their knowledge into AI workflows.

Apr 17, 2026

𝗖𝗵𝗮𝘁 𝘄𝗶𝘁𝗵 𝘆𝗼𝘂𝗿 𝗱𝗮𝘁𝗮 in Copilot, Data Agents or Databricks One? 🤖

Is your data AI ready? Are your "consumers" ready for it? Don't let AI solely write your instructions, documentation and verified answers. You need a documentation system and process to steer the process. See the guide for links to notebooks, scripts and skills that take you further! My new favorite Tabular Editor C# Macro is on slide 5 😍 I'm happy to see that if you want to succeed with Chat with your data, optimizing and documenting semantic models are still (if not more) important.

Apr 18, 2026

Building a Fabric Data Agent with On‑Premises SQL Server Data

Many organizations still rely on on‑premises SQL Server for core operational workloads — finance, billing, ERP, manufacturing, and regulatory systems — that cannot be easily migrated to the cloud. At the same time, the business is demanding: Conversational access to data (“Ask the data” experiences) AI‑powered insights without building fragile ETL pipelines Faster time‑to‑value from analytics and AI Strong governance, security boundaries, and system ownership Microsoft Fabric addresses this challenge by allowing organizations to bring AI to their data before moving their data to AI. By leveraging SQL Server mirroring into Microsoft Fabric, organizations can continuously replicate on-premises data into OneLake, making it immediately available for analytics and AI—without disrupting source systems. On top of this foundation, semantic models and Fabric Data Agents enable governed, natural-language interactions powered by Microsoft Copilot.

Apr 18, 2026

Data Agent to Data Memo

Experimented with a pipeline to explore the data autonomously, find insights, find second order signals, create a cohesive narrative that's grounded in the data & context, added an optimization layer using DSPy (Community) to tune it and converted to speech. No external APIs, services, 100% in hashtag#MicrosoftFabric . 🔊Listen, let me know. The novelty here is I used the recently published LLM-as-Verifier approach as a metric in GEPA to tune the narrative. (https://lnkd.in/g2Huycw9)

Apr 14, 2026

Fabric Ontology with Data Agent

AI built my Fabric Ontology demo in under an hour! Thanks to GitHub Copilot, I have notebooks for loading Lakehouse tables with demo data, building a Fabric Ontology, and creating instructions for a Fabric Data Agent to query the whole thing. Visit my GitHub repo to see the prompts I used, how Ontology Playground helped, and the skills files that were created along the way.

Mar 11, 2026

From API to AI Agent: Turning Raw Data into Conversational Intelligence with Microsoft Fabric

Data agent on top of Oura health data to chat with your own health data

Mar 26, 2026

What We Learned Building a Real-World Fabric Data Agent — The Honest Field Notes

The post is a detailed, field‑tested account of building a real‑world Fabric Data Agent on top of a complex industrial supply‑chain dataset, highlighting what works, what breaks, and what requires deliberate engineering. It walks through six major lessons: why agent instructions—not data sources—determine answer quality; how Semantic Models and Ontologies complement each other; the pitfalls of asynchronous graph rebuilds; the importance of using binding property names in GQL; common GQL query‑generation failures; and the hidden requirement to use dbo. prefixes in SQL. Despite the rough edges of a preview product, the author argues that Fabric IQ’s architecture—unifying SQL, DAX, and GQL behind a governed semantic layer—is powerful and worth understanding deeply for anyone building enterprise‑grade AI analytics.