Ciroos.AI, a startup that aims to automate much of the manual work involved in site reliability engineering, today said it has raised $21 million in a funding round led by Energy Impact Partners LP.

The company said it will use the funding to expand hiring and build out engineering, sales and marketing capabilities.

Ciroos describes its product as the industry’s first extensible, multi-domain artificial intelligence SRE “teammate.” The term is meant to describe an agent that supports enterprise operations teams by diagnosing and responding to system incidents across cloud, network and application layers.

Co-founder and Chief Executive Ronak Desai, who was previously senior vice president and general manager of full-stack observability and AppDynamics at Cisco Systems Inc., said the system is meant to reduce diagnosis and repair time and lower operational toil by reasoning across multiple systems and domains, which are functions traditionally handled by human SREs.

Desai said organizations use an average of 50 observability and incident response tools from vendors such as Datadog Inc., Splunk Inc., New Relic Inc., PagerDuty Inc. and ServiceNow Inc. Ciroos doesn’t look to replace those tools but to integrate with them to extract relevant data and automate aspects of root cause analysis and remediation.

The platform is built from the ground up to support multi-agent reasoning across fragmented and complex IT environments, said Desai. The SRE Teammate supports integration with existing tools via out-of-the-box connectors and the Agent2Agent protocol. It also uses the Model Context Protocol to allow its internal agents to communicate with each other. Ciroos says these components provide a flexible foundation for supporting third-party or customer-developed AI agents.

Desai said the company positions the offering not as a natural language interface on top of application program interfaces — a model many observability vendors are pursuing — but as a deeper, integrated reasoning engine that mimics human decision-making.

“Our system doesn’t just provide a UI on top of APIs,” he said. “It correlates live and historical data, connects across systems and can recommend or even execute actions such as restarting services or rolling back deployments.”

Currently, the system allows for semi-automated responses such as triggering Git-based rollbacks directly from the interface. Full automation — or “autopilot” mode — is supported where users are comfortable with autonomous remediation.

Desai co-founded Ciroos in February with Amit Patel and Ananda Rajagopal, both of whom have backgrounds in enterprise software and networking. The founding team collectively holds 84 patents in AI, observability, distributed systems, cloud, cybersecurity and networking.

The company’s business model is currently focused on direct sales to enterprises. Desai said the company is experimenting with value-based pricing models, rather than traditional volume-based or ingest-based approaches used by many observability platforms.

Image: Pixabay