Designing AI products
How we design AI products people can follow, check and trust: agent interfaces, approvals, sources and undo, and what changes when an agent can spend money. Written from shipped work with funded AI and crypto teams.
AI agent UX design: how to design an agent interface people can trust
An agent interface has to show four things a chat window hides: what the agent wants to do, what it's doing now, what went wrong and what it finished.
AI agents that move money: design rules from crypto
When an AI agent can spend money, design it like a crypto transaction.
AI product trust UX: how do you design an AI product people believe?
People trust an AI product when they can check its work.
Human in the loop design: where should an AI agent stop and ask?
An agent should stop and ask when an action can't be undone and carries a real cost: money, something sent to another person, or something deleted.

AI interface design beyond chat: what I build instead of a chat window
A chat box is the quickest AI interface to ship and one of the hardest to use well. Here's what I design instead.
No notes on that yet.