Aural - Open-source AI that conducts interviews for you

Hey Product Hunt! 👋 I'm QY, the maker of Aural. We built this because we experienced the same problem ourselves: conducting dozens of interviews is exhausting, inconsistent, and hard to scale. The idea is simple: what if AI could actually run the interview? Not just help you prep questions — but sit in the seat, ask the questions, follow up intelligently, and write the evaluation report at the end. That's Aural. You describe what you want to learn in plain language. The AI generates the interview. You share a link. Candidates show up and talk to the AI over voice, chat, or video. When it's done, you get a detailed report with scores, highlights, and areas for improvement. A few things we're proud of: - Real-time voice interviews that feel natural, not robotic - Built-in code editor (Monaco) and whiteboard (Excalidraw) for technical interviews - Anti-cheating system with tab monitoring, paste blocking, and integrity logs - Fully open source (MIT) and self-hostable — bring your own Supabase + LLM key Tech stack: Next.js 14, TypeScript, Supabase (Postgres + Auth + RLS), tRPC, and pluggable LLMs (OpenAI, Kimi, MiniMax, or any compatible API). We'd love to hear your feedback — especially around what use cases you'd want to try it for. We're actively developing and shipping fast. 3-minute demo video if you want to see it in action. https://youtu.be/Mmn1tjTzuwQ Thanks for checking it out! 🙏