Exclusive: Databricks to Acquire Tecton in AI Agent Push

Databricks is in advanced talks to acquire Tecton, a key player in the machine learning operations space, in a deal aimed at significantly bolstering its capabilities in the burgeoning field of AI agents, sources familiar with the matter confirmed on Friday.

The acquisition, which is expected to be formally announced as early as next week, would unite two of the most prominent companies in the data and AI infrastructure landscape. Tecton, a startup backed by venture capital giant Sequoia Capital, is best known for its enterprise-grade feature store, a critical component for managing the data that fuels AI and machine learning models.

Sources indicate the deal is a strategic move by Databricks to accelerate its push into the next frontier of artificial intelligence: autonomous AI agents. These agents—AI systems that can independently reason, plan, and execute complex tasks—require a highly sophisticated and reliable data foundation to function effectively in real-world business environments. Tecton's technology is designed to provide exactly that.

"This is all about building the central nervous system for AI agents," one source close to the deal told this publication. "Agents need to perceive and react to new data in real time. Tecton has built one of the best engines for that. For Databricks, this is an acquisition of core technology to power their agent strategy."

Tecton was founded by the original creators of Uber's Michelangelo machine learning platform, and their expertise in building operational ML systems at scale is highly regarded throughout the industry. Their feature and data platform allows companies to define, serve, and monitor the real-time data signals (or "features") that AI models use to make predictions. By integrating this capability directly into its Data Intelligence Platform, Databricks would be able to offer its customers a more complete, end-to-end solution for developing and deploying not just predictive models, but full-fledged AI agents.

This acquisition would also be a direct shot across the bow at competitors like Snowflake and the major cloud providers, all of whom are racing to build the definitive platform for enterprise AI. By snapping up Tecton, Databricks is not only acquiring best-in-class technology but also a team of world-class engineers with deep expertise in operationalizing AI.

While financial terms of the deal have not yet been disclosed, Tecton was last valued at over $900 million after its Series C funding round.

The move underscores a major trend in the tech industry: as the hype around foundational models matures, the focus is shifting to the "picks and shovels"—the underlying data infrastructure and MLOps tools required to make AI actually work reliably and at scale. With this acquisition, Databricks is making a bold bet that the future of enterprise AI lies not just in the models themselves, but in the data-rich operational fabric that brings them to life.

Representatives for Databricks and Tecton declined to comment.