Skip to content HVAC AI Agents home

Private beta · Energy savings

Designed to cut HVAC energy bills 10-20% — across every building you operate.

AI agents that optimize continuously against weather, occupancy, and utility tariff. ESG + LL97 reporting included. Your technicians can override — not a black box.

Join the waitlist →

10-20%

designed energy reduction

across mixed building types

no rip-out

works with existing BMS

Carrier, Trane, Daikin, Siemens, Honeywell

override-able

technicians stay in charge

every change logged, every override honored

Three modes — autonomous leads here

Primary for this page

Continuous optimization

Adjusts setpoints every 5 minutes against weather + occupancy + utility tariff + grid emissions.

Energy-manager copilot

Ask: 'which buildings missed savings target last month?' Copilot ranks + suggests next moves.

MCP for ESG tooling

Expose every kWh + setpoint to your ESG platform via MCP.

Energy savings

20–30% HVAC energy reduction — IPMVP-auditable baseline, weather-normalized, with real-time variance tracking against the plan.

HVAC AI Energy Savings establishes a per-site, weather-normalized baseline from your historical telemetry, optimizes setpoints and schedules against day-ahead grid pricing, and tracks variance so you know mid-month whether savings are on target — not after the utility bill arrives.

Most vendor-claimed energy savings are the difference between a modeled 'pre' scenario and a modeled 'post' scenario. Neither number comes from a meter. An M&V specialist or a utility rebate program won't accept that methodology, and a CFO who has seen two cycles of 'we'll save you 20%' should be skeptical of claims that can't be verified. HVAC AI Energy Savings is designed for the audience that needs to stand behind the number.

The baseline is established using your building's own historical telemetry: 12–36 months of consumption data, weather-adjusted using NOAA degree-day data specific to the building's location. The resulting baseline is IPMVP Option B-compatible — energy savings are measured at the system boundary, not estimated from whole-building models. For portfolios where Option B metering isn't available at every system, an Option C regression model is built from whole-building utility data. Both approaches produce a defensible savings claim.

On top of that baseline, the optimization layer runs continuous. Setpoints are trimmed during low-occupancy periods and pre-loaded before occupancy peaks — not on a fixed schedule, but driven by the occupancy signals your building already produces (access card events, calendar feeds, CO2 sensors). Peak shaving logic reads day-ahead grid pricing and demand-charge forecast data, then adjusts pre-cool and pre-heat windows to shift compressor load out of demand-peak windows. Demand charges represent 30–50% of a commercial HVAC electricity bill; targeting them directly is often more impactful than chasing kWh alone.

Variance tracking runs continuously against the established baseline. If a building is running 12% above its savings target at mid-month, the agent surfaces that gap with an explanation — a hot spell that exceeded the weather model, a setpoint override that wasn't reverted, a piece of equipment running inefficiently. You course-correct in the current billing cycle, not after you've already missed the target. Utility rebate exports are formatted for the major programs — Con Edison, PG&E, BC Hydro, Xcel Energy — so rebate applications submit the telemetry data the utility program actually requires.

Where it pays off

Concrete scenarios from portfolio energy leads and M&V specialists.

Four patterns we see repeatedly across CFO-driven portfolio mandates, M&V engagements, demand-charge management, and utility rebate capture.

Multi-site CFO (portfolio reduction mandate)

28-building commercial portfolio. Board has set a 25% energy cost reduction target by 2027. Facilities has tried scheduling changes and BMS upgrades but can't show auditable progress. CFO needs a number with a methodology behind it.

Energy Savings establishes IPMVP-compliant baselines across all 28 buildings, implements weather-normalized setpoint optimization, and generates a quarterly savings report the CFO presents to the board — with variance by building and cumulative progress toward the 2027 target.

22% portfolio-wide reduction in Year 1; board report generated automatically each quarter

Sustainability lead with M&V rigor requirement

Corporate net-zero commitment requires IPMVP-verified savings, not vendor estimates. Previous energy consultant delivered a 'savings report' that cited a simulation, not meter data. LEED recertification is at risk without defensible M&V documentation.

The platform's IPMVP Option B baseline isolates HVAC system-boundary metering. Savings are calculated monthly against the weather-normalized baseline, with an M&V summary report in the format required for LEED Energy & Atmosphere credit submission.

LEED M&V credit achieved; savings claim accepted without audit findings

Energy procurement / demand-charge management

Industrial campus with a 15-minute demand interval tariff. Demand charges are 42% of the monthly HVAC electricity bill. BMS has no day-ahead pricing awareness — chillers ramp up at the same time every morning regardless of grid conditions.

Peak shaving logic reads the day-ahead LMP (locational marginal price) and the campus load forecast, then pre-cools the facility 90 minutes before the predicted demand peak. Demand interval peaks drop. Monthly demand charge line falls.

18% reduction in monthly demand charges in the first 90 days

Rebate-capture for utility incentive programs

8-building portfolio in PG&E and Con Edison territory. Both utilities offer demand-flexibility rebates, but the application requires specific telemetry formats the BMS vendor doesn't export. Energy team hasn't captured a rebate in two years.

Energy Savings exports demand-flexibility event logs and baseline-versus-actual consumption data in the format required by PG&E's Commercial Demand Response program and Con Edison's AC Smart Rewards. Rebate applications submit in the same month the events occur.

$180k in utility rebates captured in Year 1 across the 8-building portfolio

FAQ

Energy savings — common questions.

  • What's the M&V methodology — IPMVP Option B or Option C?

    Both, depending on the site. Where submetering exists at the HVAC system boundary, we use IPMVP Option B: savings are measured directly from the relevant meters, not inferred from whole-building data. For sites without system-level metering, we build an IPMVP Option C regression model using whole-building utility interval data and the independent variables (weather, occupancy, production) that best explain consumption. The methodology used for each building is documented in the M&V plan produced during onboarding and referenced in every savings report.

  • How is the baseline established and weather-normalized?

    The baseline uses 12–36 months of your building's own historical consumption data — not a peer-building benchmark or vendor model. Weather normalization uses NOAA heating and cooling degree-day data specific to the building's location, with regression against your actual consumption to find the coefficients that fit your building's thermal envelope and operating profile. The resulting baseline is recalibrated if a material change occurs — a major equipment replacement, a change in occupancy, or a significant building retrofit — so the savings calculation stays accurate over time.

  • How does peak shaving work and how much does it affect demand charges?

    The peak shaving logic reads day-ahead grid pricing and your utility tariff's demand interval structure, then pre-cools or pre-heats the facility before the expected demand peak — shifting compressor load out of the peak window rather than shedding it entirely. The amount of demand charge reduction depends on your tariff structure and how spiky your current HVAC load profile is. In production deployments on 15-minute interval tariffs, we see 15–25% demand charge reductions within the first billing cycle. Demand charges typically represent 30–50% of a commercial HVAC electricity bill, so this is often the highest-ROI lever.

  • How quickly do we see savings — first billing cycle or later?

    Setpoint and schedule optimization typically shows up in the first billing cycle after optimization goes live. The improvement compounds over the first 3–6 months as the agent builds a richer occupancy model and the weather-normalization baseline stabilizes. Peak shaving savings appear in the first month where a demand-peak event occurs during an optimization window. The variance tracking dashboard shows savings in real time — you don't have to wait for the bill to confirm the trajectory.

  • Are utility rebates supported, and which programs?

    Yes. The platform exports event logs and baseline-versus-actual data in the formats required by major utility demand-response and energy-efficiency programs: Con Edison (NYC), PG&E (California), BC Hydro (British Columbia), and Xcel Energy (Colorado/Minnesota) are in production. Other utilities are added on request — if your utility publishes a rebate program with a defined telemetry submission format, we build the export. Rebate-ready reports are generated automatically after each qualifying event so your energy team isn't manually assembling documentation.

  • How is this different from a vendor-claimed percentage savings with no M&V?

    Vendor-claimed savings are typically the gap between a pre-installation energy model and a post-installation energy model. Neither number comes from a meter. The claim can't be audited, challenged, or used to support a utility rebate application, an IPMVP-required M&V submission, or a corporate sustainability disclosure. HVAC AI Energy Savings produces savings figures calculated from actual metered consumption against an established baseline, using the same methodology your utility, auditor, or sustainability certification body recognizes as defensible.

  • How is energy savings billed?

    Per building per year, with portfolio tiering. The fee covers baseline establishment, continuous optimization, M&V reporting, variance tracking, and utility rebate exports — all included. There's no performance-fee model; you pay a fixed annual amount regardless of how much you save. This matters because performance-fee arrangements create incentives to inflate savings claims. We want our numbers to be the ones that survive an audit, so the billing model doesn't reward overstating them. Contact us with your building count, metering availability, and utility territory for a specific quote.

Speaks to your existing kit

Carrier, Trane, Daikin, Mitsubishi, LG, Lennox, York, Samsung — 20+ HVAC, home-automation, and BMS brands.

63 brands across 3 categories — HVAC (31), Home Automation (18), BMS (14). Protocols: BACnet, KNX, MQTT, Matter, Modbus, REST, WebSocket, Z-Wave, Zigbee.

How it stays out of your way

Secure

Sealed data plane. Per-site auth. Audit log on every setpoint touch.

Runs on the edge

Deploys at the building edge — your data doesn't leave the site to be useful.

BYO LLM

Works with Claude, ChatGPT, and any MCP-compatible client. You pick the brain.

Private beta

See it on your portfolio first.

Designed for energy managers + ESG leads at multi-site operators.