Palmetto wants software developers to electrify America using its AI building models
If you're considering solar panels or a heat pump for your home, you might wonder, "How many panels do I need?" or "What size heat pump should I get?" Traditionally, answering these questions meant having a contractor visit your property. However, over the past decade, solar installation companies have streamlined this process significantly. Some, like Tesla, have even eliminated the need for an on-site visit altogether, thanks to advanced software solutions.
These software tools integrate various data sources to accurately predict how many solar panels a home requires. For instance, Palmetto, a solar installer, uses its own proprietary tool developed from software it acquired when it purchased Mapdwell, a solar mapping startup. But as the residential solar market growth has slowed, Palmetto has expanded its focus to include other electrification projects, such as backup batteries and heat pumps. Creating software for these indoor systems, which satellite imagery can't see, presents unique challenges.
To tackle this, Palmetto embarked on a project to compile data and develop a virtual representation of every residential building in the United States. Michael Bratsafolis, Palmetto's president of energy intelligence, described this endeavor to TechCrunch as "basically simulating digital twins of the entire U.S. residential building stock."
Their tool harnesses both public and private data. When specific data is unavailable, AI steps in to make educated guesses about what's hidden within the walls. For example, a home constructed in the 1950s likely has 2×4 lumber framing its exterior walls, which limits the amount of insulation possible.

Palmetto developed a demo to showcase some of the functionality the API can give developers access to. Image Credits: Palmetto
Bratsafolis explained that their technology can analyze a home and break it down into over 60 different characteristics and attributes. To ensure the accuracy of these models, Palmetto uses data from homes that have already purchased or leased solar panels from them. "That allows us to anonymize and use the real information and data about a home to compare how the model is performing and help train the model further," he said.
Palmetto's API and Developer Access
Palmetto has taken a bold step by opening its tool to external developers, a move inspired by Bratsafolis's previous experience at Twilio. "I came from that developer motion, the micro-service approach to providing API building blocks to kind of empower partners and developers," he remarked.
Developers can access the API with 500 free calls each month, after which it costs five cents per call. For larger clients, Palmetto is willing to negotiate customized pricing.
Bratsafolis hopes that developers will create tools that accelerate the electrification of U.S. buildings. By doing so, Palmetto will gain valuable insights into emerging market trends. "You just don't necessarily know what the use cases could be. And with an API product, you get incredible insight into the demand and the market," he said.
Related article
OpenAI to Retain ChatGPT Data Per Court Order, CEO Altman Proposes 'AI Privilege'
Many regular ChatGPT users, including the author of this article, may have interacted with the "temporary chat" feature. This option, offered by OpenAI's popular chatbot, is designed to automatically clear all information exchanged during a session a
IBM: Enterprises Deploy 'Everything' in AI, Now Face LLM-to-Use-Case Challenge
Throughout the past century, IBM has witnessed numerous technology trends emerge and fade. The approaches that endure tend to be those that offer flexibility and choice.At VB Transform 2025, Armand Ruiz, IBM's Vice President of AI Platform, outlined
Enterprise AI Costs: Claude Models Priced 20-30% Higher Than GPT in Deployment
It's widely known that different model families may employ different tokenizers. However, there has been limited research into how the actual tokenization process varies between these systems. Do all tokenizers produce the same number of tokens for i
Related Special Topic Recommendations
Comments (27)
0/500
Die KI-basierten Lösungen von Palmetto klingen vielversprechend, aber ich frage mich, wie präzise solche Modelle wirklich sind. Eine virtuelle Bewertung mag Kosten sparen, aber bei individuellen Gebäuden (Dächer, Isolierung) sind Vor-Ort-Inspektionen oft unersetzlich. Das könnte für manche Hauseigentümer ein zu großes Risiko sein. Andererseits: Wenn die Software gut genug ist, wäre das ein echter Durchbruch für die Branche. 🏠
Finde es total praktisch, dass man jetzt per KI Solaranlagen planen kann! 😊 Endlich muss nicht immer jemand vorbeikommen, um alles zu vermessen. Aber frage mich, ob das wirklich genau genug ist – bei unserem alten Haus würde ich trotzdem lieber auf Nummer sicher gehen und einen Experten konsultieren. Trotzdem, coole Sache!
Cool to see AI helping size solar panels and heat pumps! Makes me wonder if it’ll cut costs or just make contractors lazy. 😅 Anyone tried this yet?
This AI building model stuff from Palmetto sounds cool! 😎 Makes me wonder if it’ll really simplify picking solar panels or just overcomplicate things with tech jargon.
This AI building model stuff is wild! Palmetto’s making it so easy to figure out solar panels without a contractor trekking to your house. Bet it saves a ton of time, but I wonder if it’s as accurate as an in-person check. Still, super cool tech! 🌞
If you're considering solar panels or a heat pump for your home, you might wonder, "How many panels do I need?" or "What size heat pump should I get?" Traditionally, answering these questions meant having a contractor visit your property. However, over the past decade, solar installation companies have streamlined this process significantly. Some, like Tesla, have even eliminated the need for an on-site visit altogether, thanks to advanced software solutions.
These software tools integrate various data sources to accurately predict how many solar panels a home requires. For instance, Palmetto, a solar installer, uses its own proprietary tool developed from software it acquired when it purchased Mapdwell, a solar mapping startup. But as the residential solar market growth has slowed, Palmetto has expanded its focus to include other electrification projects, such as backup batteries and heat pumps. Creating software for these indoor systems, which satellite imagery can't see, presents unique challenges.
To tackle this, Palmetto embarked on a project to compile data and develop a virtual representation of every residential building in the United States. Michael Bratsafolis, Palmetto's president of energy intelligence, described this endeavor to TechCrunch as "basically simulating digital twins of the entire U.S. residential building stock."
Their tool harnesses both public and private data. When specific data is unavailable, AI steps in to make educated guesses about what's hidden within the walls. For example, a home constructed in the 1950s likely has 2×4 lumber framing its exterior walls, which limits the amount of insulation possible.

Bratsafolis explained that their technology can analyze a home and break it down into over 60 different characteristics and attributes. To ensure the accuracy of these models, Palmetto uses data from homes that have already purchased or leased solar panels from them. "That allows us to anonymize and use the real information and data about a home to compare how the model is performing and help train the model further," he said.
Palmetto's API and Developer Access
Palmetto has taken a bold step by opening its tool to external developers, a move inspired by Bratsafolis's previous experience at Twilio. "I came from that developer motion, the micro-service approach to providing API building blocks to kind of empower partners and developers," he remarked.
Developers can access the API with 500 free calls each month, after which it costs five cents per call. For larger clients, Palmetto is willing to negotiate customized pricing.
Bratsafolis hopes that developers will create tools that accelerate the electrification of U.S. buildings. By doing so, Palmetto will gain valuable insights into emerging market trends. "You just don't necessarily know what the use cases could be. And with an API product, you get incredible insight into the demand and the market," he said.
OpenAI to Retain ChatGPT Data Per Court Order, CEO Altman Proposes 'AI Privilege'
Many regular ChatGPT users, including the author of this article, may have interacted with the "temporary chat" feature. This option, offered by OpenAI's popular chatbot, is designed to automatically clear all information exchanged during a session a
IBM: Enterprises Deploy 'Everything' in AI, Now Face LLM-to-Use-Case Challenge
Throughout the past century, IBM has witnessed numerous technology trends emerge and fade. The approaches that endure tend to be those that offer flexibility and choice.At VB Transform 2025, Armand Ruiz, IBM's Vice President of AI Platform, outlined
Enterprise AI Costs: Claude Models Priced 20-30% Higher Than GPT in Deployment
It's widely known that different model families may employ different tokenizers. However, there has been limited research into how the actual tokenization process varies between these systems. Do all tokenizers produce the same number of tokens for i
Die KI-basierten Lösungen von Palmetto klingen vielversprechend, aber ich frage mich, wie präzise solche Modelle wirklich sind. Eine virtuelle Bewertung mag Kosten sparen, aber bei individuellen Gebäuden (Dächer, Isolierung) sind Vor-Ort-Inspektionen oft unersetzlich. Das könnte für manche Hauseigentümer ein zu großes Risiko sein. Andererseits: Wenn die Software gut genug ist, wäre das ein echter Durchbruch für die Branche. 🏠
Finde es total praktisch, dass man jetzt per KI Solaranlagen planen kann! 😊 Endlich muss nicht immer jemand vorbeikommen, um alles zu vermessen. Aber frage mich, ob das wirklich genau genug ist – bei unserem alten Haus würde ich trotzdem lieber auf Nummer sicher gehen und einen Experten konsultieren. Trotzdem, coole Sache!
Cool to see AI helping size solar panels and heat pumps! Makes me wonder if it’ll cut costs or just make contractors lazy. 😅 Anyone tried this yet?
This AI building model stuff from Palmetto sounds cool! 😎 Makes me wonder if it’ll really simplify picking solar panels or just overcomplicate things with tech jargon.
This AI building model stuff is wild! Palmetto’s making it so easy to figure out solar panels without a contractor trekking to your house. Bet it saves a ton of time, but I wonder if it’s as accurate as an in-person check. Still, super cool tech! 🌞





Home






