Launched this week

CueLoop (limited early access)
Real-time AI coaching for better user interviews
13 followers
Real-time AI coaching for better user interviews
13 followers
Running user interviews is hard. You miss follow-ups, ask leading questions, and only find the gaps when you listen back. CueLoop is a macOS app that coaches you during interviews in real time. It captures both sides of the conversation, transcribes live, and delivers gentle nudges through a discreet overlay near your Mac's notch. It catches bias, suggests follow-ups, and tells you when you're talking too much. A world-class UX researcher in your corner. Free during limited early access.





Just checked out the website, first of all, the infinite animation in the landing page looks 🔥
I think you found a blue ocean here! I don't think I saw any products that do what you claim to do.
I will point out 1 important thing here: not everyone uses Mac, so I would advise implementing a web version or at least supporting Windows also.
Hey everyone! I'm Brent. I've spent over 10 years in product management, and the one thing I've learned is: the quality of what you build is directly tied to the quality of your customer/user conversations.
But most builders aren't trained interviewers. We ask leading questions we don't notice, miss follow-ups, and gloss over emotional cues. Other AI tools try to solve this by replacing you with an AI interviewer entirely, which is a terrible trade off in my opinion. CueLoop doesn't take you out of the room. It makes you better in it.
I built CueLoop to coach you during interviews. It captures both sides of the conversation, transcribes live, and nudges you when something matters. All through a discreet overlay near your Mac's notch that aims to be subtle and not distracting.
It's in limited early access right now (50 available sign ups, 3 sessions per user). I'd love feedback, especially if you run customer interviews yourself. What would make this useful for you? I would love to chat!
Love the idea of lightweight coaching layers for this kind of use case!