.png){kind=logo}
From Chicago’s hyperscale corridors to data centers anchoring Virginia’s sprawling Dulles Technology Corridor, the demand for AI compute infrastructure has made mission-critical facility development one of the most consequential disciplines in the built environment. The projects are large, the regulatory environments are varied, and the stakes, both financial and operational, leave little room for energy compliance errors. Over his tenure at Stantec, Senior Building Analyst Sabhya Katia has led the sustainability, energy modeling, and compliance work on more than 20 data center projects spanning the continental United States, with individual project capacities ranging from 12 MW to 108 MW. The cumulative scale of that portfolio, and the consistency of its outcomes, positions Katia among the practitioners who have most directly shaped how mission-critical facilities are designed and approved at a national level.
Katia’s project portfolio reaches across six of the country’s most active data center markets: Chicago, Texas, California, Virginia, Atlanta, and Charlotte. Each market presents its own regulatory character. Illinois operates under the IECC 2021 energy code framework. California requires compliance with Title 24 through CBECC, the state’s own building energy compliance tool, which imposes requirements that diverge meaningfully from the ASHRAE-based national standard. Virginia and the mid-Atlantic corridor have their own AHJ conventions and review processes. In Texas and the Southeast, the density of new hyperscale development has created permitting environments under constant pressure from the volume of applications.
Across all of these environments, Katia has served as the lead responsible for thermal analysis, energy modeling, compliance documentation, and stakeholder coordination from concept through permit acceptance. The projects he has led are not uniform: they include air-cooled facilities, liquid-cooled deployments, and increasingly, mixed-plant designs that combine both cooling modalities in a single facility. Each configuration presents distinct compliance challenges. The energy modeling for a facility that blends air-cooled and liquid-cooled zones requires normalization approaches that standard tools and standard reviewers are not always equipped to handle. Katia has led that work on every project in the portfolio.
A 12 MW facility and a 108 MW hyperscale campus are not the same engineering problem. At smaller deployments, the challenge in energy compliance is precision: the efficiency gains that liquid cooling or optimized air systems deliver are real, but at lower absolute power consumption levels, the noise in the modeling data is proportionally larger and harder to defend in a permit submission. Demonstrating an 18% improvement over the ASHRAE baseline requires tighter documentation and more granular component-level evidence when the facility’s total load is measured in single-digit megawatts.
At hyperscale, the challenge shifts. A 108 MW campus involves a scale of plant equipment, redundancy configurations, and cooling infrastructure where transient simulation, modeling system behavior not just under steady-state conditions but through startup sequences, load changes, and fault events across all 8,760 hours of the year with varying temperature, humidity, and wind conditions, becomes both technically demanding and regulatorily essential. Permitting authorities reviewing a facility of that size require confidence not just that the design performs well under ideal conditions, but that it behaves predictably under the full range of operating scenarios it will encounter over a decades-long facility life.
What Katia’s portfolio demonstrates is consistency of methodology across that range. The same rigor applied to a 12 MW deployment in Charlotte applies to an 108 MW campus in Northern Virginia, adapted for scale and jurisdiction, but built on the same analytical foundation. That consistency is what allows owners and developers working with Stantec’s Midwest and Atlantic teams to anticipate a predictable path from concept to approved permit, regardless of the project’s size or location.
One of the underappreciated complexities of national-scale data center development is that the United States does not have a single energy code. Individual states and counties adopt, adapt, and amend model codes, IECC, ASHRAE 90.1, Title 24, on their own schedules and with their own modifications. A compliance approach that works cleanly in Illinois may require significant restructuring for California, not because the underlying physics changed, but because the regulatory framework for demonstrating compliance is different.
Across 20+ projects, Katia has navigated this patchwork directly, adapting documentation and modeling approaches to each jurisdiction’s specific requirements while maintaining the analytical consistency that makes the work defensible. For California projects, that has meant working within CBECC’s framework for a building type, the high-density liquid-cooled data center, that the tool’s original developers did not anticipate. For IECC-jurisdiction projects, it has meant constructing the code-equivalence arguments that closed liquid loop systems require when reviewers have no prior submission to reference.
The data centers Katia has helped design, model, and permit are not abstract infrastructure. They are the physical substrate of the AI workloads that enterprises, cloud providers, and research institutions are deploying at accelerating rates. A facility certified to LEED BD+C: Core & Shell standards and accepted under state energy code is a facility that can attract the institutional tenants and capital that modern data center development requires. A facility that fails compliance review, or that reaches the permit stage without adequate energy modeling documentation, faces delays that cascade through project timelines and capital schedules.
Across more than 20 projects and six major data center markets, Katia has delivered permit-ready documentation and certified energy models that have allowed those facilities to proceed on schedule. The aggregate deployable compute capacity represented by that portfolio, projects ranging from 12 MW to 108 MW, translates directly into AI infrastructure that would otherwise be delayed or unbuilt. In an industry where capacity constraints are already shaping which AI applications get deployed and when, the practical consequence of that work extends well beyond the projects themselves.
“Every project is different, different code, different jurisdiction, different cooling configuration,” Katia says. “But the core question is always the same: can you prove, to a standard a regulator will accept, that this facility performs the way you say it does? Getting that right is what determines whether the project moves forward.”
