1. ARCHITECTURAL INTELLIGENCE — AI Trained to Understand the Platform
Other platforms are starting to connect AI to their APIs. The AI reads and writes data. That's it. A fancy form filler.
We did something completely different. We trained the AI to understand FrameworX at an architectural level — the four pillars, UNS as the backbone, ISA-95, module independence, the build sequence. The AI builds a mental model of the platform before it ever touches a configuration. It can give architectural guidance. It can explain design decisions. It can recommend best practices — not because someone wrote a rule table, but because it internalized the platform logic.
It's not autocomplete. It's a colleague that truly understands FrameworX.
2. FULL SOLUTION COVERAGE — AI Knows Every Object in Your Application
When people hear "AI in SCADA," they think the AI creates displays. Or writes scripts. Or helps with tags. One thing.
Our AI knows ALL objects. Tags, alarms, historian, devices, protocols, databases, scripting, reports, displays, symbols, dynamics, layouts — everything. The entire solution.
Now think about what that really means. A customer has an application built 5 years ago by an engineer who left. Nobody understands it. With our AI, they open that solution and in 10 minutes they have the full picture — what's configured, how things connect, what the logic does, where potential issues are. Documentation, auditing, impact analysis, security review, maintenance — all enabled now.
This is not a development tool. This is a platform intelligence tool. Creating new solutions is just the beginning.
3. LIVE IDE INTEGRATION — The Designer and AI Act as One
This one still amazes me every time I use it.
The AI and the Designer share context in real time, both ways. Everything the AI creates appears instantly on the Designer screen — live. And the other way too: the engineer can be looking at any configuration and ask the AI "what is this doing?" — the AI already sees what the Designer is showing. No copy-pasting, no exporting, no explaining what screen you're on.
From the user's perspective, the LLM and the Designer act as one entity. Two interfaces — one visual, one conversational. You use whichever is faster for what you need in that moment.