Georgia Tech Startup Targets Building Energy Inefficiencies With AI and Drones

Making buildings environmentally sustainable is a major thrust of Lamarr.AI. The startup leverages advances in drones, thermal imaging, and AI to autonomously detect building defects. Georgia Tech Associate Professor Tarek Rakha co-founded Lamarr.AI to apply this research and scale it across the U.S. and the globe. (Photo credit: Wes McRae, Georgia Tech)

Buildings are big energy consumers, emitting wasteful carbon, contributing to a warming planet, and accounting for more than 40% of the world’s carbon dioxide emissions.

The culprits behind this hefty energy hit are two distinct sources of carbon: the building construction process and a building’s operation and use.

Fortunately, a new startup is paving the way for dramatic reductions in carbon from building operations. Lamarr.AI has perfected the process of using drones, thermal imaging, and artificial intelligence (AI) to diagnose the health of building exteriors and roofs.

The young company — named for Hedy Lamarr, the actor and inventor credited with helping birth the modern telecommunications era — offers an autonomous solution to diagnose a building’s energy footprint. Lamarr.AI is a spin-off from Georgia Tech, Massachusetts of Technology (MIT), and Syracuse University.

 

A Lamarr.AI drone flies over a building. (Photo credit: Tarek Rakha)

“Our service is akin to giving your building an MRI scan. It will revolutionize health monitoring for the built world,” said Tarek Rakha, Lamarr.AI CEO and associate professor in Georgia Tech’s School of Architecture. “We actually diagnose the root cause of a building envelope’s defects, map them out on 3D models, and translate them to an energy simulation to quantify lost energy.”

According to the Department of Energy, the U.S. wastes $100 billion in energy due to ineffective building envelopes, which are the shell, or barrier, between a building’s interior and the outdoors.

An expert on computer-vision drones, urban energy sensing, and computation informatics, Rakha currently directs Georgia Tech’s High Performance Building Lab. He collaborated on Lamarr.AI’s development and commercialization with Syracuse University’s Senem Velipasalar and MIT’s Norhan Bayomi and John E. Fernández. Together, the three universities co-own two patents based on Lamarr.AI’s technology.

In October 2024, the company raised $1.1 million in pre-seed funding to bring its building diagnostics technology to market. Hazelview Ventures, the venture capital arm of global real estate firm Hazelview Investments, led the funding round. Other investors include milemark•capital, Newlab, Georgia Tech Foundation Ventures, and MIT’s SBXi Fund. 

To date, Lamarr.AI has worked with significant players in the healthcare, higher education, multi-family, residential office building, and industrial warehouse sectors to help building owners identify targeted, cost-effective repairs that save energy.

“I’m very happy with the response we’ve gotten as a very young startup — already making revenue,” said Rakha. He attributes this positive reception to Lamarr.AI’s solution being “cheaper, faster, more accurate, and safer” than the regular means of inspecting a building’s envelope.

Michael Williams, Hazelview Investments managing partner and head of development, said that Lamarr.AI first piqued his interest because of its use of AI and its hyperfocus on building envelopes, which are costly and time-consuming to inspect.

“Having the ability to automate that process and make it a little bit easier on a national scale was extremely compelling to us,” Williams said. “It’s going to have a tremendous impact and will be a game-changer if they continue to evolve the product.”

Lamarr.AI’s leadership team includes researchers from Georgia Tech, MIT, and Syracuse University. Pictured on the MIT campus (L to R): John E. Fernández, CSO and co-founder; Ben Rocci, head of strategy and business Development; Tarek Rakha, CEO and co-founder; Norhan Bayomi, COO and co-founder; and Senem Velipasalar, CTO. (Photo courtesy of Lamarr.AI)

Data from early users has shown potential energy savings as high as 30%, depending on the building’s age.

“The return on investment can be met in two to five years and brings benefits to building occupants in terms of more comfortable living conditions,” said Rakha. Buildings that implement recommended fixes enjoy better indoor air quality, as well as lower energy bills.

Lamarr.AI has already helped building owners save nearly $1 million in engineering and construction costs by automatically identifying problems like deteriorated window seals and leaky roofs, in addition to helping effectively plan and scope large building-envelope retrofits.

According to Rakha, his company’s autonomous approach differs dramatically from current energy audits or inspection processes. These are performed manually using a handheld thermal camera to pinpoint problem areas, which isn’t feasible for larger structures.

There are also safety considerations. Traditionally, a comprehensive inspection of a skyscraper, such as the Empire State Building in New York City, would require people “scaling it like Spider-Man,” he explained. Instead, Lamarr.AI flies drones away from the building — a faster process that doesn’t endanger human lives.

As drones fly around a building, they “see” evidence of thermal leakage and other defects. In the wintertime, for example, the drone might pick up a defect that means the building is leaking heat; the AI then identifies the root causes. The company’s AI can also detect cracks and structural risks in building exteriors. When combined with thermal defects, this creates a comprehensive analysis of the building envelope’s health and performance — building managers, architects, and other urban development stakeholders can then act on this information.

The building sector will play a key role in the fight against climate change, with adoption of net-zero building certifications on the rise. Lamarr.AI offers a cost-effective, fast path to reducing a building’s carbon footprint and environmental impact.

Rakha emphasized that the race to help buildings achieve carbon net-zero goals is urgent. “There is no other option,” he said, given the rapid pace of the climate crisis.

This research was supported by a Department of Energy BENEFIT 2018 Aerial Intelligence for Retrofit Building Energy Modeling (AirBEM) award.