AI-powered interview prep: adaptive MCQs, live coding with instant feedback, and a system-design whiteboard scored like a senior engineer. Find a bug, collaborate with friends for challenges
No reviews yetBe the first to leave a review for PrepyAI
Maker
📌
Hi Product Hunt 👋 I'm Tomer, the founder.
I built PrepyAI because interview prep tools kept handing me problems instead of practice at the actual interview. An interview is really about how you reason out loud — and that's hard to practice alone.
So PrepyAI recreates the interview across three formats: adaptive MCQs, live coding with instant AI feedback, and a system-design whiteboard that scores your architecture against senior-level criteria. Everything feeds a dashboard so you can see exactly where you're weak.
It's free to start, no card. I'd love your honest feedback — especially on the grading. Where does it feel accurate, and where does it miss? Happy to answer anything about how it works under the hood. AMA!
Report
How does the senior engineer scoring for the system design whiteboard actually work in practice, like is it comparing my answer against some rubric or is it a more open ended LLM judgment?
Report
Maker
@mcahitkratc17y Hey :) Happy to share -> It's both. Every question comes with its own rubric - the checklist you see in your feedback. The AI reads your whiteboard like an interviewer would (components, connections, and what your labels say) and marks each rubric item as met, partial, or missed, citing the part of your design that earns it. The score itself isn't the AI's opinion - it's calculated directly from that checklist.
How does the senior engineer scoring for the system design whiteboard actually work in practice, like is it comparing my answer against some rubric or is it a more open ended LLM judgment?
@mcahitkratc17y Hey :)
Happy to share -> It's both. Every question comes with its own rubric - the checklist you see in your feedback. The AI reads your whiteboard like an interviewer would (components, connections, and what your labels say) and marks each rubric item as met, partial, or missed, citing the part of your design that earns it. The score itself isn't the AI's opinion - it's calculated directly from that checklist.