Menu
standards

Do We Really Need Standards for Demand Flexibility?

No, we don’t!

Why? Because when devices describe themselves, their capabilities, and properties, you can fine-tune an LLM to understand what you mean without you having to spell it out. 

Examples:

“What’s the cabin temp?”
→ Cabin Temperature for Batmobile is 72°F
(It knows you mean your Tesla’s cabin. Because you have one!)

Make it multi-device:

“How much charge do I have, how many miles can I drive, and what’s the price?”
→ State of Charge: 44.72%
→ Range: 111.2 miles
→ Utility Price (oadr3): $0.0864

Not just energy:

“Is it paired? And what’s playing?”
→ PulseSound: Paired (Bluetooth speaker)
→ YouTube Player: Classical

Stack properties and commands:

“Turn on both Hues, then set left Hue to red in 3 seconds and right Hue to blue in 4.”

Even automation & optimization:

“If price < $0.75 and range > 150 miles, set charge limit to 90% and start charging. But only on weekdays.”

Add a plugin platform for any device or service, and standards become irrelevant. Plugins describe themselves in English, and it just works because the platform understands what you mean!

That’s NuCore.ai: an MIT-licensed (FREE) fine-tuned LLM and natural-language platform for Smart Buildings. Run it locally at the edge (with llama.cpp) or in the cloud.

The age of standards is ending.

The age of understanding has begun. Come play with it today!

Smart,Home,With,Blue/light,Green,Glassmorphism,Our,Vision,Who,We

Is It Paired?

The short answer is Yes! Because it knows you’re talking about your Bluetooth speaker.

So why exactly do we still need mappings, ontologies, and standards just to adjust thermostat set points, charge a car, optimize DERs (Distributed Energy Resources), or even pair Bluetooth speakers or play music?

The short answer is: we don’t.

With NuCore.ai, a fine-tuned LLM and platform for Smart Buildings, MIT licensed and able to run locally at the edge or in the cloud, none of that is needed.

Things describe themselves in plain English at runtime. You speak to it as you would a person, and the model understands what you mean.

  • “What’s the cabin temp?”
    → Cabin Temperature for Batmobile is 72°F (it knows you’re asking about your Tesla’s cabin temp, because you have one!)

  • “How much charge do I have and how many miles can I drive and what’s the price?”
    → State of Charge for Batmobile is 44.72%
    → Estimated Range for Batmobile is 111.2 miles
    → Price for Utility Signals (oadr3) is: 0.0864$

  • “Turn on both Hues and then set left Hue to red in 3 seconds; set right Hue to blue in 4 seconds.”

  • “If price < $0.75 and miles I can drive is more than 150, set charge limit to 90% and start charging. But only on weekdays.”

The age of standards is ending.

The age of understanding has begun, and you are invited to play with everything right now!

Indoor,Photo,Of,Dramatic,Image,Of,An,Artistic,Looking,Computer

From OPEn to NuCoreAI

It’s hard to believe it’s only been seven months since we introduced OPEn. It started as a bold idea: a plugin framework that uses AI to understands smart home and IoT devices, interpret their properties and commands at runtime, and generate intelligent execution plans for optimization. In essence, eliminating the need for rigid standardization across device and grid protocols.

By February 2025, we had a fully functional demo running on cutting-edge tech: Claude 3.5 Sonnet, RAG with Pinecone, and a healthy dose of optimism. The results were stunning, the interest was immense, and we were flying high … until we realized we were spending over $100/month on AI services for just one demo! At roughly 10 queries a day over two months, reality set in: the business model simply wasn’t sustainable.

And that’s when the real adventure began: the birth of NuCoreAI and our deep dive into the fascinating world of local AI. Suddenly, our days were filled with fine-tuning, parameter counts, LoRA/PEFT, Alpaca-style datasets, vector databases, context window gymnastics, Hugging Face, Unsloth, Colab sessions, temperature tweaking, GPUs vs. NPUs, and yes, prompt engineering (on loop).

Somewhere in that whirlwind, we had more than a few LOL moments, especially when testing prompts that explicitly told LLMs not to make things up. Naturally, they responded with absolute confidence, and zero regard for accuracy:

Me: What is nodeshit?
LLM: Nodeshit is a command that allows you to interact with IoX nodes directly from the IoX shell.

Me: What is crap?
LLM: In the context of IoX, ‘crap’ refers to a device that is not working properly or at all. It can also describe a device causing issues or conflicts with the IoX system.

So where are we now? At a pretty exciting milestone: we’re seeing up to 80% accuracy using LLaMA 3 8B, with reranking as a first step. Next up: exploring model and technique optimizations, such as caching, and defining minimum hardware requirements for maximized accuracy and speed.

Want to dive in? Check out our GitHub for the code, models, and all the AI goodies we’ve cooked up so far.

CEC Davis

2025 California Demand Flexibility Summit

,

Join us May 22nd for the 2025 California Demand Flexibility Summit.

We will have a table with NuCoreAI platform optimizing a few cool things such as Tesla, Ecobee, and Philips Hue using OpenADR 3.0 Price signals!

Location

UC Davis Conference Center

550 Alumni Lane Davis, CA United States, 95616