Microsoft acquires Seattle startup Osmos to automate data engineering in fabric

Microsoft announced the acquisition of Osmos, a Seattle-based startup specialising in agentic AI for data engineering. The technology and the Osmos team will join the Microsoft Fabric engineering organisation to develop autonomous data processing capabilities within Microsoft’s unified analytics platform.

Osmos, founded in 2019 by former Microsoft and Google engineers Kirat Pandya and Naresh Venkat, focuses on simplifying the data lifecycle. The startup’s tools use large language models (LLMs) to automate the ingestion, cleanup, and transformation of raw data. Microsoft plans to embed these "data agents" directly into Microsoft Fabric to reduce the manual burden faced by data teams.

“Many teams spend most of their time preparing data instead of analysing it,” wrote Bogdan Crivat, Corporate Vice President of Azure Data Analytics, in an official blog post. Crivat noted that the acquisition aims to turn raw data into "AI-ready assets" within OneLake, the central data lake for Fabric.

The acquisition follows a 2024 partnership where Osmos launched an "AI Data Wrangler" specifically for Microsoft Fabric. This tool allowed users to clean unruly data, such as messy Excel files or PDFs, without writing code or scripts. By internalising this technology, Microsoft aims to offer native autonomous agents that work alongside users to map schemas, reconcile data types, and generate production-ready pipelines.

As part of the integration, Osmos began winding down its standalone product suite this month. This includes its Datasets, Pipelines, and specialised Data Agents for external platforms like Databricks. Current Osmos customers received notification that these services would begin sunsetting in January 2026 as the team shifts focus toward Fabric’s core infrastructure.

Osmos previously raised USD 13 million in a 2021 Series A funding round led by Lightspeed Venture Partners, with participation from CRV and SV Angel. While the financial terms of the acquisition were not disclosed, public records indicate Osmos operated with a small, specialised engineering team of fewer than 20 employees.

This move marks a shift in Microsoft’s data strategy toward "agentifying" the enterprise. By embedding agentic AI at the infrastructure level, Microsoft intends to automate routine data maintenance, allowing organisations to connect and share data with significantly lower operational overhead.