Work Categories
What we can do for you
Ideas and insights
Why shouldn't the AI-generated results only be "re-generated"? Editing, partial modification and version comparison are the true collaborative experiences
To what extent can AI agents perform automatically? Truly high-risk actions must be designed with "confirmation, preview and revocation".
Why do AI answers need to display the source? Citation UX truly addresses the question "Can I verify this sentence?"
How should the input box of an AI product be designed? The Prompt Box should not make all users start from scratch
How can form error prompts be written so as not to drive users crazy? "Format error" is not a qualified solution
How to choose between Toast, Banner and in-site message? Notification design is not about "informing users", but about managing attention
Why is the permission design of enterprise software the most feared? The Role/Permission UX should let users know "who can see what"
Is it necessary to pop up a second confirmation before deleting? The design of hazardous operations should be determined based on whether they can be revoked or not
Why are SaaS backend tables getting more and more difficult to use as they are made? The real problem with the data table is not "too much information", but that the tasks have no priority
Why don't complex products showcase all their functions at once? Progressive disclosure is not about "hiding things", but about controlling cognitive costs
Building useful, scalable digital products around user experience