How OpsMill’s Infrahub Turns Fragmented Infrastructure Data Into an AI‑Ready Source of Truth

Enterprise IT environments generate huge volumes of infrastructure data, but much of it lives in disconnected systems—spreadsheets, configuration management databases (CMDBs), scripts, and vendor tools. That fragmentation makes the data inconsistent and difficult to trust, which in turn blocks automation and creates risk for AI‑driven operations.

Paris‑based startup OpsMill is trying to solve this problem with Infrahub, an open‑source platform designed to act as a unified, versioned system of record for infrastructure data across networks, data centers, and cloud environments. The company raised $14 million in Series A funding led by IRIS, with participation from BGV, Serena, and Partech, to expand the platform and grow its engineering and product teams.

The Core Problem: Fragmented Infrastructure Data

Many organizations attempt to automate IT operations using scripts, orchestration tools, and infrastructure‑as‑code workflows. But those tools depend on accurate infrastructure data. In practice, that data is often scattered across different systems and updated inconsistently.

When the underlying information is unreliable, automation becomes risky—and AI systems can make decisions based on incomplete or outdated context. This “dirty data” problem has become a major barrier to adopting AIOps and AI‑driven infrastructure management.

OpsMill’s approach is to consolidate that fragmented information into a single, structured data layer that both humans and automation systems can trust.

Infrahub: A Unified Source of Truth for Infrastructure

Infrahub is designed as a schema‑first, versioned platform for infrastructure data management. Instead of storing infrastructure details across disconnected tools, organizations model their networks, data centers, and cloud resources within one unified data model.

The platform acts as a central repository where infrastructure data is validated, structured, and synchronized across systems. By consolidating technical and operational information in one place, teams can maintain consistency across automation tools, configuration systems, and operational workflows.

This unified data model is intended to eliminate the inconsistencies that frequently cause automation failures.

Why Infrahub Uses a Graph Database

A key design choice in Infrahub is its graph database architecture. Infrastructure environments are inherently complex networks of relationships—devices connect to circuits, services depend on clusters, and policies apply across multiple layers of infrastructure.

Graph databases are particularly well suited to modeling these relationships. They allow systems to represent how infrastructure components connect and depend on one another instead of storing isolated records.

This relational view gives engineers—and increasingly AI systems—a deeper operational context. For example, automation tools can evaluate how a proposed configuration change might affect dependent services before applying it.

Native Version Control for Safer Automation

Infrahub also incorporates built‑in version control for infrastructure data, similar to Git workflows used in software development.

Changes to infrastructure models can be:

• proposed and reviewed
• compared with previous versions
• validated through automated checks
• safely rolled back if problems occur

This workflow helps teams validate updates before they affect production systems, reducing the risk of automation errors.

Version control also provides a clear audit trail, which is critical when organizations begin allowing automated systems—or AI agents—to recommend or execute operational changes.

Building an AI‑Ready Data Layer

OpsMill positions Infrahub as a foundational layer for AI‑driven infrastructure operations.

AI agents rely heavily on structured, trustworthy data to reason about systems and take action. If the infrastructure data they consume is fragmented or inconsistent, their decisions can quickly become unsafe.

By unifying and structuring infrastructure data, Infrahub creates a consistent operational context that AI systems can interpret. This makes it easier for enterprises to experiment with AI‑assisted automation while maintaining oversight and reliability.

What the $14M Funding Round Signals

OpsMill’s $14 million Series A round, led by IRIS with participation from BGV, Serena, and Partech, reflects increasing investor interest in platforms that prepare enterprise infrastructure for AI‑driven operations.

The funding is expected to support:

• continued development of the Infrahub platform
• expansion of engineering and product teams
• broader go‑to‑market efforts

Early traction suggests enterprises are already exploring the model. The platform is reportedly used in production at organizations including TikTok, while a European cloud provider has said it reduced deployment times from five days to fifteen minutes after adopting the system.

The Bigger Shift: Infrastructure Data as the Automation Foundation

OpsMill’s strategy highlights a broader shift happening in enterprise IT. Rather than focusing solely on automation tools or AI models, many organizations are realizing that data quality and structure are the real prerequisites for automation at scale.

Without a trusted system of record, automation pipelines and AI agents can amplify errors rather than eliminate them. Platforms like Infrahub aim to fix that underlying layer by making infrastructure data consistent, auditable, and machine‑readable.

If that approach gains traction, the infrastructure stack may evolve in the same way software development did—where version‑controlled systems of record became the foundation for safe, large‑scale automation.