Introducing EnergyPilot AI

AI-powered tools for the building retrofit industry

The building retrofit market is entering a period of unprecedented growth. Between IRA incentives, state and local decarbonization mandates, and rising energy costs, the economic case for commercial building retrofits has never been stronger. The pipeline of potential projects is enormous.

The bottleneck isn’t demand. It’s the front end of the sales process.

Before an ESCO can write a performance contract, before an engineering firm can scope a project, before an equipment manufacturer can size a solution — someone has to answer a set of deceptively simple questions. How much energy does this building actually use? Where is it being wasted? Which upgrades will deliver the most savings? And can we estimate those savings reliably enough to justify pursuing the project?

Today, answering these questions for a single building requires weeks of engineering analysis. Answering them across a portfolio of hundreds of buildings — the scale at which most ESCOs and A&E firms need to operate — is effectively impossible with current tools. Teams end up relying on rules of thumb, spreadsheet heuristics, and engineering judgment to decide which opportunities to pursue. Good projects get overlooked. Resources get spent chasing projects that don’t pencil out. The screening process becomes the constraint on growth.

EnergyPilot AI is built to solve this.

We’re an AI-powered platform for building retrofit sales enablement. Our tools help ESCOs, engineering firms, and equipment manufacturers evaluate retrofit opportunities faster and with more confidence — from first look at a utility bill through preliminary savings estimates and ECM selection.

EnergyPilot AI combines building energy forecasting models trained on hundreds of thousands of commercial building profiles with a calibration system that incorporates a building’s actual utility data to produce building-specific predictions. The result is a tool that can take incomplete billing data — the 5 to 8 months of utility bills you typically have at the start of an engagement — and produce a credible full-year energy baseline with calibrated uncertainty bounds.

From that baseline, EnergyPilot AI evaluates energy conservation measures and estimates savings across the building types, climate zones, and fuel types that the commercial retrofit market covers. It’s designed to give your team a defensible first look at a project in minutes rather than weeks — not to replace the investment-grade audit, but to ensure that engineering resources are deployed on the opportunities most likely to deliver.

We come to this problem with deep domain experience. EnergyPilot AI is built by the team behind nZero’s energy data platform, with years of experience in building energy modeling, carbon accounting, and machine learning for the energy sector. We’ve spent the last several years working with large-scale building energy datasets, developing forecasting models, and learning — from real customer engagements — where the gaps are between what the industry needs and what existing tools provide.

The biggest gap we found was at the beginning of the process: the screening and qualification stage where decisions are made with the least information. That’s where we’ve focused.

Over the coming weeks, we’ll be publishing a series of posts that go deeper into our approach:

• How we handle the sparse billing data problem that makes early-stage baselining so unreliable with traditional methods
• A practitioner’s guide to understanding when standard methods break down — and what to do about it
• How our models move from baseline to retrofit evaluation, including how we think about uncertainty and the sim-to-real gap

We’re also working with pilot partners in the ESCO and equipment manufacturing space to validate our approach against real-world retrofit outcomes. If you work in the retrofit industry and the problems we’ve described here resonate, we’d welcome the conversation.

EnergyPilot AI is building AI-powered tools for the building retrofit industry. For more information, visit EnergyPilot AI.