OnCue
Your AI-powered interview wingman & application mentor.
Core trick
Transcribes live interview audio, detects questions, and runs vector search over a personal experience vault to surface matching stories.
Description
Your AI-powered interview wingman & application mentor. OnCue listens to interview questions in real-time and surfaces relevant memories from your experience vault exactly when you need them. OnCue also gives you great suggestions on what to write about in your written applications through an MCP vector search tool.
README excerpt
# OnCue Your AI-powered interview wingman. OnCue listens to interview questions in real-time and surfaces relevant memories from your experience vault exactly when you need them. [](https://www.youtube.com/watch?v=4UaQEB1b84E) ## What It Does OnCue uses live audio transcription to detect interview questions and automatically pulls up the most relevant information from your personal knowledge base using vector search. Think of it as having all your best stories, projects, and accomplishments on standby—appearing right on cue. **The Magic Flow:** 1. Start recording during your interview 2. Interviewer asks: "Tell me about a challenging project..." 3. OnCue detects the question and searches your memory vault 4. Relevant experiences appear instantly with similarity scores 5. Answer confidently with context at your fingertips ## Tech Stack - **Frontend**: React 19 + TypeScript + Vite + Tailwind CSS - **AI/ML**: Google Gemini (audio transcription, embeddings, LLM detection) - **Backend**: Node.js + Express - **Database**: MongoDB Atlas with Vector Search - **Models Used**: - `gemini-2.5-flash-native-audio-preview-09-2025` (live audio transcription) - `text-embedding-004` (vector embeddings) - `gemini-2.0-flash-exp` (optional LLM-based question detection) ## Features ### Real-Time Audio Transcription - Captures microphone input and streams to Google Gemini - Live, sentence-level transcription updates - D