i’m building kiwi (try-kiwi.com): an ai design critique tool that reviews your prototype (figma / screenshots / links) using configurable “personas” + heuristics, and outputs actionable issues + suggested fixes. the goal is to catch usability problems fast, before you ship.
i’m trying to validate if this is actually useful for builders (not just designers).
product demo: https://www.youtube.com/watch?v=n-Q09jvXUTI&t=70s
quick questions (pick any):
- what’s your current workflow for catching usability issues pre-ship? (friends, usertesting, ship + pray, etc.)
- where do you get stuck: “what to build next”, “copy/ux clarity,” “layout/a11y,” “flows,” “edge cases”?
- would you use an automated critique if it was: (a) fast, (b) specific, (c) tied to your target user, (d) not generic?
- what output format would you want: checklist, annotated screenshots, prioritized bug list, jira tickets, diffs, prompt you can paste back into cursor/v0?
- what would make you not trust it? (false positives, vague advice, inconsistent personas, etc.)
- pricing expectation: $0 / $10 / $30 / $100+ per month? or pay-per-prototype?
if you’re down to test: reply with what you’re building + what you’d want critiqued (landing page, onboarding, pricing, dashboard, etc.).
thanks, trying to avoid building in a vacuum.