Most AI companies today are building extraordinary engines. GPT-5 vs Claude. Benchmark scores. Token speeds. And forgetting to design the cabin. This gap is what we call the Experiential Architecture problem.
"How was your flight?" I asked a friend who had just returned from Dubai on Emirates. She talked for 11 minutes. The seat. The chicken option that was actually good. The air hostess who smiled like she meant it. The way the cabin felt quiet even though there were 400 people in it. She did not mention the Rolls-Royce Trent 1000 engine powering the aircraft. She never noticed it. That was the point.
In AI today, everyone is in a race to build a better engine. GPT-5 vs Claude. Benchmark scores. Token speeds. Context windows. We write about it; we debate it; we post comparison tables on LinkedIn :) And the person actually using the product? The patent attorney. The radiologist. The financial analyst. They are asking exactly one question: "Does this feel like it understands what I need?" Not: "what model is running underneath?"
"I made this mistake myself. Early builds of eety were technically impressive. The invention understanding was deep; the claim logic was sound; the outputs were accurate. I was proud of it. And then a patent attorney told me: 'it doesn't feel like it gets me.' That sentence cost us 3 months and a complete rethink of the product surface."
Experiential Architecture starts from the opposite end. You begin with one question: what does the expert feel at the exact moment they use this tool? What would make them feel heard, capable, in control... like the tool was built specifically for how they think? And then you work backwards to figure out which technology makes that feeling possible.
The engine becomes invisible. On purpose. The best products in the world work exactly this way. You do not know the compression algorithm your music app uses; you just know the song started instantly. You do not know which LLM is reading your patent disclosure; you just know it understood the invention better than you expected. That is the goal. The engine should never be the story. The experience should never be an afterthought :)
Tabrez explains Experiential Architecture
The Emirates story, the cabin problem, and what it means for AI product design. ~4 minutes.
Video in production · Coming soonSome articles are about the philosophy; the industry-wide pattern. Others are about specific product decisions in eety, seen through the EA lens.
Philosophy Track
Industry thinking · Broader audience
The engine vs the cabin.
These articles don't require a patent background. They are about a pattern visible across all AI products; and what it would look like if builders started from feeling instead of features.
First article coming soon.
Articles will appear here as they publish.
Practice Track
eety decisions · Patent attorneys
What EA looks like in the product.
Each article maps one specific felt moment in eety to the Experiential Architecture principle behind it. If you use the product, you have already experienced these; this series names them.
Anything goes at check-in
Multi-file upload · Any format, any combination
The departure board was always there
Understanding Elements · Visibility into what the AI learned
The airport was not built to make you wait
Background processing · Your time is not the tool's waiting room
They brief before every flight, without exception
90% Threshold · Preparation always comes before movement
They'll wrap whatever you bring
Patent Drawings · Rough sketch, your own, or generated from scratch
You chose your seat. The output should sound like you chose it too.
Style DNA · The output should sound like your drafting voice
The airline remembered you. So does eety.
Matter Memory · Memory is respect for the relationship
Every bag has a tag. Every claim should too.
Visual Cascade · Traceability from claim to source material
The connecting flight always knows where the bag is going
Cross-section coherence · One document, one shared understanding
These are not features. These are feelings. EA is the discipline of designing for them; then choosing the technology that makes them possible.
"
The moment the tool stops itself and says: I don't think I have enough to draft this well yet. Can I ask you three things first?
The 90% Threshold
"
The moment the output sounds like something you would have written on your best day. Not like AI wrote it. Like you did.
Style Intelligence
"
The moment you return to a matter three weeks later and the tool remembers everything. Not just the draft; the decisions, the reasoning, the direction you chose.
Matter Memory
"
The moment you say "start over" and it does. Without argument. Without losing what was actually good. Just: how do you want it done?
Editable by Design
Five applications, no credit card. If it doesn't feel like it gets you in the first twenty minutes; it probably isn't ready for you yet. That's a promise we make because we mean it.