Veuillez entrer vos informations personnelles pour vous abonner au blog Microsoft Fabric.

Merci de vous être abonné au blog Microsoft Fabric. Vous recevrez des notifications par courrier électronique lorsque nous publierons de nouveaux messages.

Microsoft Fabric Updates Blog

  1. Mises à jour
  2. Data Science

Enhance data prep with AI-powered capabilities in Data Wrangler

All Fabric notebook users have access to Data Wrangler, a code-generating tool with an immersive interface for exploring and transforming pandas or Spark DataFrames. Data Wrangler provides a library of common data-cleaning operations that you can browse and apply seamlessly, getting real-time previews and generating reusable code. 

Thanks to new AI-powered capabilities in Data Wrangler, including automated suggestions and custom operations with Copilot, you can now do even more to accelerate exploratory analysis and data preparation in Fabric. 

Get automated suggestions with rule-based AI 

Unsure where to start? A new set of automated suggestions analyzes your data with rule-based AI from the Microsoft PROSE team, highlighting the most relevant Data Wrangler operations for you. Every time you apply a new operation, the suggestions refresh based on the new state of your data. 

New AI-powered suggestions highlight Data Wrangler operations that are relevant to you.

Ask Copilot to apply custom operations 

Need an operation that you don’t see in Data Wrangler? You can now use Copilot to generate custom code. All you need to do is describe the desired operation. As with any Data Wrangler step, you get a preview before applying or discarding the code.  

With Copilot in Data Wrangler, you can generate custom code from your own instructions.

Translate custom code from pandas to PySpark

Working on big data? Data Wrangler converts Spark DataFrames to pandas for performance reasons, then translates all exported code back to PySpark. With GenAI in Data Wrangler, custom code is also translated to PySpark—whether you type it in yourself or generate it with Copilot. 

Data Wrangler can translate all your operations from pandas to PySpark, using AI where necessary to produce the most accurate code.

Next steps

We’re eager for you to try out the new capabilities in Data Wrangler and let us know what you think.

  • If you’re interested in using Data Wrangler to apply the new set of Fabric AI functions—for tasks like sentiment analysis, entity extraction, and custom text generation—sign up for the preview experience with this link: aka.ms/DataWranglerAIPreview. 

Billets de blog associés

Enhance data prep with AI-powered capabilities in Data Wrangler

Image vedette du blog

Introducing Copilot for Real-Time Dashboards: Write KQL with natural language
janvier 21, 2026 par Michal Bar

Turning questions into KQL queries just became part of Real-Time Dashboard tile editing experience, using Copilot. This new feature brings the power of AI directly into the tile editing workflow. When editing a tile, you’ll now see the Copilot assistant pane ready to help you turn natural language into actionable queries. Whether you’re new to … Continue reading “Introducing Copilot for Real-Time Dashboards: Write KQL with natural language”

En savoir plus
Image vedette du blog

Create Embeddings in Fabric Eventhouse with built-in Small Language Models (SLMs)
janvier 8, 2026 par Adi Eldar

What if generating embeddings in Eventhouse didn’t require an external endpoint, callout policies, throttling management, or per‑request costs? That’s exactly what slm_embeddings_fl() delivers: a new user-defined function (UDF) that generates text embeddings using local Small Language Models (SLMs) from within the Kusto Python sandbox, returning vectors that you can immediately use for semantic search, similarity … Continue reading “Create Embeddings in Fabric Eventhouse with built-in Small Language Models (SLMs)”

En savoir plus