New Repertoire: What product designers need to know to work with AI

Original title: Novo Repertório: o que product designers precisam saber para trabalhar com IA

AI tools now draft, generate and decide inside the design process. Designers who treat them as faster versions of old tools end up reviewing output they can’t evaluate. This book gives designers the working knowledge to use AI with judgement: what these systems do well, where they break, and how to keep control of the result.

It moves from the uncertainty AI brings into UX work to the practices that contain it: giving AI the right context, deciding what to delegate, reviewing what comes back, and writing specifications and acceptance criteria that AI-generated work has to meet.

Inside the book:

✦ AI UX and behavioural systems.

✦ Invisible friction in agentic interactions.

✦ Automation boundaries and trust.

✦ Human oversight in AI-driven decisions: when an agent acts, and when it should ask.

✦ Decision-making in AI-driven products.

✦ Multi-step coordination: making agent logic visible to the people it affects.

✦ Uncertainty in UX work with AI

✦ Context and continuity

✦ Deciding, delegating and reviewing

✦ AI in the designer’s work

✦ AI tools and execution quality

✦ A conclusion and an AI glossary for designers.


Printed edition with a poster.

(free digital edition on Bold)

VIEW DETAILS

Intelligent Interfaces: Designing AI Experiences, Behavioral Systems and Product Decisions

Original title: Interfaces Inteligentes: projetando comportamento, decisões e limites em produtos com inteligência artificial

In an AI product, the interface is more than the screen. It’s how the system behaves. AI gives fluent answers through inference, and those answers can be wrong or biased while sounding certain. UX methods built for predictable flows don’t cover this.

This book is about designing that behaviour. It treats the designer as someone who sets the operating and ethical limits of a probabilistic system, then keeps reviewing them after launch.

What it covers:

✦ Behaviour design for AI systems, beyond screens and flows

✦ Language as interface: prompts as the decision architecture that guides how the AI reasons

✦ Showing uncertainty: confidence levels that help people decide instead of trusting blindly

✦ When an AI agent should act and when it should ask

✦ Automation limits, human oversight and trust

✦ Making an agent’s logic visible to the people its decisions affect


Coming soon!

GET LAUNCH UPDATES