Groundcover has raised $100 million in Series C funding led by One Peak as the observability market shifts its focus from traditional post-production monitoring to the infrastructure requirements of AI and agentic systems. The round brings Groundcover's total funding to $160 million and comes roughly 15 months after the company closed a $35 million Series B in April 2025. Groundcover was founded in 2021 and builds its observability platform on eBPF and OpenTelemetry, two open-source technologies that have become central to collecting runtime data in modern cloud environments.

CEO and co-founder Shahar Azulay framed the funding as evidence of a broader change in how engineering teams think about observability. "I think what is happening to observability right now is fascinating," Azulay said in an interview. For most of the past decade, observability spent its time as a post-production discipline, catching outages and cutting the time engineers need to find a root cause. That focus is now shifting as agentic AI systems move into the software development lifecycle, pulling production context earlier into coding, testing, and deployment work.

The evolution of observability

Observability has traditionally been concerned with three pillars: metrics, logs, and traces. Together, these data types help engineers understand how a system behaves in production, why a request failed, or why an application is running slowly. Over time, the discipline grew from simple monitoring to more complex distributed tracing, especially as organizations adopted microservices and containerized infrastructure. But the underlying assumption was always that human engineers would be the ones reading the telemetry and making decisions.

The rise of AI agents is changing that assumption. Agentic workflows can generate thousands of decisions per minute, invoke external tools, rewrite code, and interact with production systems in ways that are difficult to predict. As a result, observability platforms must now handle not only traditional application data but also AI model behavior, token usage, and the path an agent takes from prompt to action. That is a very different problem from catching a failed database connection.

What eBPF does and why it matters more now

Groundcover's core approach is to eliminate the need for developers to instrument every service by hand. The platform uses eBPF, short for extended Berkeley Packet Filter, a Linux kernel technology that lets code run safely inside the kernel without a custom kernel module. eBPF has long been used for network monitoring, but Groundcover applies it to application and infrastructure observability. According to Azulay, that removes a step most observability vendors still require. "You didn't have to have the developer instrument an SDK, change their code base, and so on."

The same property is becoming useful for a different reason now. Engineering teams are adopting new AI tools fast enough that they lose track of what is actually running in their own environment, Azulay said. He compared the gap to the visibility problems teams dealt with roughly a decade ago, before observability tooling matured. Because eBPF operates below the application layer rather than depending on code a developer wrote, Groundcover can still see workflows nobody thought to instrument. "eBPF is kind of that security net of even if you didn't instrument, even if you're not in full control, you're gonna know which agentic workflows are running in production, which models are using, which vendors they're using, and so on."

eBPF's role has expanded as organizations adopt AI-assisted development tools. The kernel-level visibility it provides is particularly useful in AI infrastructure because it can capture telemetry from the entire stack, including network traffic, system calls, and application activity, without requiring changes to application code. In an environment where AI agents are spawning new processes and making unpredictable calls, this kind of blanket visibility can help engineers understand what the system is actually doing rather than what they expected it to do.

How agentic workflows are breaking distributed tracing

Distributed tracing has long been a cornerstone of observability. It follows a request as it moves across services so engineers can see where time is spent and where something broke. That approach relies on a predictable number of hops, the kind of path an engineer could trace by hand, such as a cache calling a database. Azulay said that assumption breaks down once agents enter the picture, since a single agent session can generate a large number of tool calls and internal model calls with no fixed pattern. "With LLMs and agentic workflows, this is becoming very complicated."

The complexity is not just about scale. Teams now also track token usage and hallucination rates alongside traditional metrics like latency and error rate, Azulay said. Traces can contain a customer's actual prompt instead of only structured request data, which raises privacy questions. Azulay does not consider the result a variant of application performance monitoring. "It's not going to be the same product. AI observability is not exactly APM."

Privacy and telemetry storage

Groundcover's architecture is designed around the idea that AI workloads produce more telemetry and more sensitive telemetry. The platform stores telemetry inside the customer's own cloud environment rather than a shared vendor backend. That approach, Azulay said, lets customers hold the larger, more sensitive telemetry volumes agentic workloads produce without shipping that data to a third party. "I think people are expected to save more telemetry, and save more telemetry more privately."

This is a departure from the traditional SaaS observability model, where customer data is centrally aggregated by the vendor. It also aligns with a growing enterprise requirement around data governance and AI compliance. Companies running AI models often have strict policies about where prompts and model outputs can be stored, and an observability platform that copies that data into a vendor environment can create an immediate compliance problem. By keeping telemetry in the customer's VPC or cloud account, Groundcover believes it can offer a more secure alternative.

Agent Mode and MCP integration

Groundcover is not just optimizing its platform for the needs of modern agentic AI activity. It is also using AI to improve user experience. Agent Mode is Groundcover's built-in AI assistant for engineers, letting them ask questions about their systems, build dashboards, and troubleshoot problems in logs and traces without writing queries by hand.

The company has also built a Model Context Protocol (MCP) integration that connects Agent Mode to coding agents and workflow tools including Linear, allowing engineers and AI systems to pass context back and forth during an incident. MCP is emerging as a standard way for AI models to interact with external tools and data sources, and its adoption has grown as more organizations deploy AI coding assistants. Azulay said Groundcover's MCP integration has moved faster than expected. Customers use the integration differently depending on how far along they are in adopting AI tools, according to Azulay. Some ask questions through it instead of opening the Groundcover dashboard, while others use it to write a fix directly.

He framed the pattern as an industry trend rather than a product win specific to Groundcover. "People are basically starting to build their autonomous software development structure," he said. This suggests that observability tools are becoming part of a larger feedback loop in which AI agents not only write code but also monitor the systems they build.

A growing funding round

The $100 million Series C was led by One Peak, with participation from Morgan Stanley Expansion Capital, Zeev Ventures, Angular Ventures, Heavybit, and Jibe. The new investors' participation shows that the observability market is still attracting serious venture capital attention, even as many enterprise software categories face funding pressure. Groundcover's total funding now stands at $160 million, and the company is headquartered in Tel Aviv, Israel.

The company's growth is tied to the evolution of software development itself. For years, observability was largely about production monitoring: catching problems after code shipped and helping engineers trace them to their source. The rise of AI agents is changing that equation. Because agents can make decisions in production, generate code, call external APIs, and interact with user data, they require a new set of monitoring capabilities. The same eBPF-based approach that eliminated manual instrumentation gives Groundcover a path into this emerging market.

As agentic AI becomes more embedded in enterprise software, the demand for observability platforms that can handle unstructured, high-volume telemetry is expected to grow. Groundcover's bet is that by combining eBPF technology with private telemetry storage and AI-native tools, it can capture a meaningful share of that market. The company is also betting that developers will increasingly want to interact with their observability data through natural language and AI assistants rather than through dashboards and query languages.

Groundcover at a glance

• Founded: 2021
• Total funding: $160 million
• Latest round: $100 million Series C, led by One Peak
• Other investors: Morgan Stanley Expansion Capital, Zeev Ventures, Angular Ventures, Heavybit, Jibe
• Headquarters: Tel Aviv, Israel
• CEO: Shahar Azulay
• What they do: Observability technology built on eBPF and OpenTelemetry

Source:Network World News