Ecolab embraces agentic AI and streamlines root-cause analysis with Azure SRE Agent

Ecolab’s site reliability teams were inundated with about 30 daily performance alerts, manually navigating multiple dashboards to identify and fix root causes.

The company was eager to streamline operations by embracing Azure SRE Agent, using agentic AI to triage incidents across diverse data sources and recommend precise remediation steps in close collaboration with Microsoft.

Within months, the daily alert volumes dropped to less than 10 on average. The SRE team now resolves issues faster and focuses on optimizing performance rather than putting out fires.

“My vision for the SRE team is to empower them to focus on what truly matters, reducing the time spent on determining root causes. This is where I believe the SRE Agent holds tremendous potential, paving the way for a more efficient and impactful workflow.”

Don Patrichi, Software Engineering Manager, Ecolab

Ecolab is pioneering agentic AI to drive bold innovation. These autonomous systems go beyond traditional automation—streamlining workflows, uncovering insights, and accelerating decisions. By delegating complex tasks to AI agents, the company aims to boost efficiency, personalize customer experiences, and unlock new revenue streams.

Embracing agentic AI for site reliability operations

For over 100 years, Ecolab has been at the forefront of developing science-backed solutions to enhance food safety, maintain clean landscapes, and optimize the use of vital resources. To advance its critical mission, Ecolab is pushing the boundaries of agentic AI to transform reliability operations.

The company’s site reliability engineering (SRE) teams are responsible for detecting, analyzing, and remediating incidents that affect solution reliability and performance. “We have a very complex environment with approximately 550 Microsoft Azure resources in just one of our architectural monoliths, and it’s probably the largest in terms of notification alerts,” says Don Patrichi, Software Engineering Manager at Ecolab.

Before adopting agentic AI, Ecolab’s SRE teams had to triage tens of daily alerts across the company, investigating the causes and how to resolve them—a time-consuming process as they navigated multiple dashboards and data sources.

“We have a very complex environment with approximately 550 Microsoft Azure resources in just one of our architectural monoliths, and it’s probably the largest in terms of notification alerts.”

Don Patrichi, Software Engineering Manager, Ecolab

Streamlining incident reports and root-cause analysis

In August 2025, the company made a bold decision to transform its SRE operations as an early adopter of Microsoft Azure solution Azure SRE Agent, an innovative, pre-built SRE solution that automates repetitive tasks and improves incident response. Ecolab worked closely with Microsoft to co-build the product through feedback and experimentation, tying SRE Agent into multiple data sources in Ecolab’s infrastructure.

For example, the company brought alerts from Azure Monitor into SRE Agent, which also presents Ecolab employees with specific remediation steps to be considered. Ecolab moved quickly to incorporate the SRE Agent with Azure DevOps, and the company worked closely with Microsoft to automate the creation of tickets alerting developers to specific incidents.

As incidents are flagged, Ecolab uses prompt engineering to engage SRE Agent in a direct dialogue, empowering teams to uncover root causes more quickly. “Azure SRE Agent changes our overall interaction with the system, allowing us to pose questions and receive precise, targeted answers,” says Patrichi.

“Azure SRE Agent changes our overall interaction with the system, allowing us to pose questions and receive precise, targeted answers.”

Don Patrichi, Software Engineering Manager, Ecolab

Shaping the future of agentic AI and reducing daily alerts to less than 10

In just a couple of months of running the preview, Ecolab has streamlined site reliability operations. The number of daily notification alerts has plummeted from 30-40 to less than 10 on average, and Ecolab receives a daily report summarizing incidents and assessing the system’s overall performance. With those tools and the ability to interact directly with SRE Agent, SRE teams spend less time on root-cause analysis and lay the foundation for deeper automation in the future.

The company’s close collaboration with Microsoft offered Ecolab a unique opportunity to request new features and services in SRE Agent, and these are now shaping the final product. One engineer asked that SRE Agent alerts go to a communication and collaboration service like Microsoft Teams for instant visibility. “Working with Microsoft isn’t just about adopting a tool—it is about shaping it,” says Patrichi. “The product team is responsive to inputs, and we influenced key features, like the consolidated daily report, which are now finely tuned to meet our needs.”

Laying the groundwork for autonomous incident resolution

With its embrace of agentic AI, Ecolab is advancing toward a future where responding to operational incidents requires minimal manual effort, a revolutionary transformation for its SRE teams. The company is actively exploring how to use the GitHub Copilot Coding Agent to link incidents with underlying source code, further streamlining incident response and allowing teams to focus on building resilient operations.

“My vision for the SRE team is to empower them to focus on what truly matters, reducing the time spent on determining root causes,” says Patrichi. “This is where I believe the SRE Agent holds tremendous potential, paving the way for a more efficient and impactful workflow.”

Discover more about Ecolab on Facebook, Instagram, LinkedIn, X/Twitter, and YouTube.