How is your enterprise tracking AI agent telemetry? Groundcover thinks it should never leave your cloud

The AI agent observability space is rapidly expanding, with a new entrant named Groundcover raising $100 million in funding led by One Peak. The company reports having over 250 paying customers, a tripling of annual recurring revenue in the past year, and is increasingly replacing established observability platforms within enterprise environments. This surge in momentum comes as the AI agent observability market gains ground against industry giants like Datadog, Dynatrace, New Relic, Splunk, and Grafana.

Groundcover argues that the rise of AI systems has fundamentally altered the assumptions underlying traditional observability platforms. Rather than focusing on features, the company seeks to convince enterprises that the underlying architecture of observability must evolve alongside AI systems, which are becoming more autonomous and generating vast amounts of telemetry.

By viewing telemetry as an infrastructure problem, Groundcover aims to address the growing need for complete operational context in AI-driven environments.

The proliferation of AI-assisted software development has led to an unprecedented increase in telemetry, including prompt execution, model latency, token consumption, retrieval pipelines, tool invocations, and agent behavior. As organizations embrace AI agents that execute complex workflows and interact with production systems, the value of retaining comprehensive telemetry becomes increasingly apparent.

However, traditional pricing models based on data ingestion volume pose challenges, as engineers often resort to sampling traces or limiting data retention to mitigate costs, inadvertently reducing visibility which is crucial for understanding AI system behavior.

Groundcover's solution involves a bring-your-own-cloud (BYOC) architecture, where customers retain control of telemetry storage and processing within their own cloud environments, such as AWS, Azure, or Google Cloud. This approach eliminates the need for Groundcover to charge based on telemetry ingestion, instead pricing customers based on the size of their monitored infrastructure.

The company believes this shift in pricing strategy encourages customers to retain complete telemetry for operational analysis, compliance, and AI-assisted troubleshooting, rather than opting for data sampling to cut costs. While other observability vendors have introduced AI-powered features, Groundcover contends that the true game-changer lies in rethinking the architecture itself, rather than merely adding AI capabilities on top of conventional SaaS platforms.