I maintain AI-Mock-Interview-Platform, a TypeScript project on my @ud-ai GitHub account, github profile. It is a full-stack, voice-first mock interview platform for dynamic, back-and-forth conversations between candidates and an AI interviewer.
My AI-Mock-Interview-Platform repository was Started 2026-06, last updated 2026-06.
The platform drives conversational transitions using stateful LangGraph logic and Google Gemini.
My frontend stack is Next.js 14 (App Router) + TypeScript + Tailwind CSS.
The client state uses Zustand (Client).
The database and session cache stack is PostgreSQL (via Prisma ORM) + Redis (Session cache).
The AI stack is LangGraph + Google Gemini (1.5 Flash & 1.5 Pro).
The voice SDK is Vapi.ai WebRTC (with free native Web Speech API fallback).
My five-command quick start is npm install; cp .env.example .env.local; npx prisma migrate dev; docker run -d -p 6379:6379 redis:alpine; npm run dev.
The README says to fill in keys after copying .env.example to .env.local. It does not specify the environment-variable names or required keys, nor provide further details on the interview flow, LangGraph state design, Gemini model selection, database schema, or Vapi.ai setup.