Quick Facts
- Groundcover raised $100 million in Series C funding led by One Peak, bringing total funding to $160 million at a reported $500 million valuation.
- The company tripled annual recurring revenue and doubled its workforce to 140 employees in the past year, with more than 250 paying customers.
- Its bring-your-own-cloud architecture stores telemetry inside a customer’s own AWS, Azure, or Google Cloud environment, not on vendor-managed infrastructure.
Groundcover has raised $100 million in Series C funding to expand its observability platform for engineers and autonomous AI agents. One Peak led the round, with participation from Morgan Stanley Expansion Capital and existing investors Zeev Ventures, Angular Ventures, Heavybit, and Jibe. The deal values the company at $500 million, according to CTech.
The raise follows a year in which groundcover tripled annual recurring revenue and doubled its global workforce. The company now has 140 employees across offices in Israel, Boston, and San Francisco. It counts more than 250 paying customers, including a Fortune 5 enterprise, and closed several seven-figure deals in the past twelve months.
The Architecture Bet
Groundcover’s core product uses a bring-your-own-cloud model. Customers run the data plane, including telemetry storage and processing, inside their own cloud accounts. Groundcover provides a managed control plane and user interface on top. Telemetry stays in cloud object storage such as Amazon S3, which costs roughly $0.023 per gigabyte per month.
That pricing matters. Competitors like Datadog charge per host, per gigabyte of logs, and per million spans for application performance monitoring. A comparable setup at medium scale can exceed $50,000 per month on those platforms.
The company’s timing looks deliberate. Six weeks before the funding announcement, Datadog abandoned its SaaS-only model and adopted the bring-your-own-cloud architecture that groundcover and a handful of rivals had championed. Groundcover’s lead investor read that move as confirmation. “The AI era demands a completely different observability architecture, and groundcover is the only company that has built it,” said David Klein, co-founder and managing partner at One Peak.
AI Agent Telemetry Is the New Problem
AI applications produce a different class of operational data than traditional services. Prompt execution, model latency, token consumption, retrieval pipelines, tool invocations, and agent behavior all generate telemetry that engineers need to interpret. For enterprises running autonomous systems, that data is also sensitive.
Most AI-assisted observability tools route production logs, traces, and prompts through external large language models, which means that data leaves the customer’s environment. Groundcover’s new Agent Mode runs on a model service inside the customer’s own cloud account. The feature is compliant with GDPR and CCPA by architecture, not by policy, and carries no AI surcharge. Customers pay their own model provider token costs directly.
CEO Shahar Azulay said the volume problem predates AI but has grown sharper. “All existing solutions are built on the same principle: you need vast amounts of data to understand how your code is performing,” he said at a recent media briefing. “The volume of information keeps growing, but in practice companies pay much more without getting proportionally greater value.”
The Technology Stack
Groundcover uses eBPF, a Linux kernel technology, to watch application and infrastructure activity without requiring developers to instrument each service manually. The 2026 CNCF Observability Technical Advisory Group survey found that 67 percent of teams running Kubernetes at scale had already adopted at least one eBPF-based observability tool in production.
The platform captures AI telemetry from two sources. eBPF auto-detection provides immediate per-call visibility with no code changes. OpenTelemetry instrumentation adds full agent-level data for teams that want it. Telemetry volumes in modern production environments are doubling every few months, and AI workloads are accelerating that rate.
What This Means for Buyers
Groundcover’s pitch to enterprise buyers is straightforward: keep your data, cut your bill, and meet your compliance requirements without policy workarounds. The company is positioning Agent Mode as a way to bring AI into production operations without sending sensitive operational data to a third party.
Azulay said the funding will go toward expanding the platform’s agentic capabilities and growing its sales motion in North America. “We plan to change the way engineers and agents collaborate to bring true autonomy into production operations,” he said.