Context

Glenart Group is a boutique data center development management and owners’ representative firm that helps clients take data center projects from site selection through construction and into operations.

During the operations phase, one of their key deliverables is creating the operational documentation facility teams need to safely run data centers, including maintenance and emergency procedures.

As data center projects grow in complexity, the volume of required documentation increases significantly. A single site can require hundreds of MOPs, SOPs, and EOPs, traditionally created manually by facilities engineers and commissioning technicians.

The Problem

Before DOME, creating operational procedure libraries was a highly manual and time-consuming process.

Each MOP, SOP, and EOP was written individually by facilities engineers or commissioning technicians. For a single data center site, this could mean creating hundreds of procedures manually, requiring significant engineering time dedicated to documentation production instead of facility operations.

The process typically took 8 to 12 weeks per site and did not scale. Every new client project required teams to repeat the same manual documentation workflow, creating a growing challenge as the number of data center projects increased.

Objectives

Glenart Group wanted to build DOME to transform the manual process of creating data center procedure libraries into a scalable software-driven workflow.

The goal was to create a platform that could ingest a validated asset inventory along with real project documentation, including OEM manuals, submittals, and design documents, and use AI to generate asset-specific procedures grounded in the site's actual information.

The objective was not to create generic AI-generated content, but a reliable system that could help produce accurate operational documentation based on each facility’s unique equipment and documentation.

Solution

To solve this challenge, we built DOME — an AI-powered platform designed to automatically generate data center operational procedures based on real project documentation.

The platform ingests validated asset inventories and site-specific documentation, including OEM manuals, submittals, and design documents. Using AI-powered document processing and retrieval, DOME generates procedures that are grounded in the facility’s actual equipment and information rather than generic AI-generated content.

We built the complete platform from the ground up, including the frontend, backend, database, document ingestion and retrieval pipeline, and AI integrations.

The system was developed with 13 different modules covering areas such as authentication, asset management, document ingestion, and procedure generation, delivering a complete workflow for creating asset-specific procedures.

Results

DOME transformed the way Glenart Group approaches operational documentation for data center projects.

Previously, creating a single asset-specific procedure could take around 4 hours of manual work. With DOME, the same procedure can now be generated in a few minutes, while still referencing the site's actual equipment and documentation.

The platform changed documentation from a manual engineering cost center into a scalable service offering that Glenart Group can provide to clients.

Beyond speed improvements, DOME created a repeatable process that allows Glenart Group to generate procedure libraries more efficiently without repeating the same manual documentation effort for every new client site.

Key Learnings

The success of DOME highlighted that AI systems create the most value when they are built around real business workflows, trusted data, and domain-specific knowledge.

For engineering and industrial processes, generic AI output is not enough. The system needs to understand the organization’s actual documents, assets, and processes to produce reliable results.

DOME demonstrated that by combining Glenart Group’s industry expertise with AI-powered automation, a highly manual and time-intensive process can be transformed into a scalable software solution.