Generative UI: interfaces built on the fly
Interfaces where the LLM doesn't just chat: it produces components, forms, views and dashboards adapted to the user's intent and the context of the moment.
The problem we solve.
Traditional interfaces assume you know in advance every screen your product needs. Generative UI assumes the opposite: the model decides at each interaction which components to present, in what order and with what data, based on what the user needs right now.
It's an emerging pattern that changes the rules of product design. Done well, it removes unnecessary screens and drastically reduces friction. Done badly, it generates inconsistent interfaces that confuse the user.
What is included
- A documented, consistent catalogue of generatable components
- Rules for when the model may create UI vs when it must request it
- Fallback patterns when the generated UI is invalid
- Validation system for visual and brand coherence
- User tests on the predictability of the generated interface
Our process for Generative UI.
Design
We define the building blocks the model can compose and the constraints that keep coherence.
Integration
We connect the component catalogue with the model (function calling, JSON Schema, structured outputs).
Iteration
We collect usage feedback to expand the catalogue or restrict it where necessary.
Cases where it adds real value
Personalised analytical dashboards that adapt to the user's role. Complex product configurators where the next step depends on the previous one. Internal assistants that generate different forms depending on the ticket. The common rule: when the space of possible screens is huge and designing them all by hand would be absurd.
Other services within AI-native experiences
AI product strategy
How to integrate AI into your existing product or build a new one with AI at the centre.
View service →Functional AI prototyping
Functional prototypes connected to real LLMs to validate use cases in days, not months.
View service →AI agent design
Designing the agent experience: how they present themselves, escalate to humans and manage uncertainty.
View service →Context engineering
Designing the context that feeds the models: prompts, tools, memory, RAG and usage policies.
View service →Multimodal UX
Experiences that combine text, voice, image and video naturally within the same flow.
View service →