Snowflake Acquires Observe to Enhance Observability Capabilities

Snowflake has announced its plans to acquire Observe AI, a well-funded startup known for its namesake observability platform.

The financial terms of the deal have not been disclosed. When acquisition talks first emerged last month, The Information reported, citing sources, that Observe was valued at approximately $1 billion—more than two and a half times the amount the nine-year-old company had raised from investors.

Observe's platform enables developers to identify root causes behind application slowdowns, website outages, and other technical issues. It also captures data related to routine aspects of a company’s technology operations. For instance, engineering teams can use Observe to monitor inference costs generated by large language models.

Snowflake intends to integrate Observe’s observability tools into its cloud-native data platform. A key focus of this integration will be AI SRE, a component developed by Observe specifically for site reliability engineers.

As the name suggests, AI SRE is an artificial intelligence-powered chatbot designed to assist developers in diagnosing technical incidents. Users can not only ask the AI to pinpoint the source of a problem but also customize how it performs analysis. For example, AI SRE can be instructed to examine only error logs from the past 24 hours when investigating a newly detected issue.

AI SRE is powered by a data management engine called the O11y Context Graph. According to Observe, this system ingests hundreds of terabytes of telemetry data daily from customer environments. It links relevant data points, creates indexes—essentially shortcuts that accelerate search—and generates materialized views, which are cached query results that reduce loading times.

Observe stores processed data within Snowflake’s environment. The acquisition is expected to streamline troubleshooting by enabling customers to extend their data retention periods.

A retention window refers to the duration organizations keep archived telemetry data before deletion. Historically, companies often purged such data to control storage expenses—a practice that could complicate incident investigations. For example, if a system failure originated two months ago but logs are only retained for one month, engineers may struggle to trace the root cause.

Snowflake reduces infrastructure costs by compressing data and storing it in low-cost object storage services. The company claims its object storage capabilities will allow clients to maintain telemetry records significantly longer than typical setups, providing more context during incident reviews.

"An organization's ability to remain resilient is fundamentally limited by how much data it can ingest, how effectively it navigates disparate formats and silos, and how quickly it can reason across that data," said Christian Kleinerman, Executive Vice President of Products at Snowflake, in a blog post. "Together, Snowflake and Observe will empower customers to manage enterprise-wide observability with a modern, scalable architecture—enabling them to run production workloads and agents with greater confidence, without compromising data for cost control."

This acquisition follows Snowflake’s previous purchase just two months ago. In mid-November, the company agreed to acquire Datometry, a firm that built tools for migrating data from competing platforms into Snowflake. Datometry also developed OpenDB, an open-source relational database.