Summary

This open-source AI interview platform, Lingwu, tackles hiring inefficiencies by automating candidate screening and structured interviews. It empowers HR, recruiters, and institutions to conduct voice, text, or video interviews, generating questions, conversations, and analysis reports. The platform significantly reduces repetitive tasks, ensures consistent evaluation, and offers a cost-effective solution for scaling hiring workflows, also serving as a practice tool for job seekers.

How It Works

Built with Next.js 14, tRPC, and React/Tailwind CSS, it uses Supabase for Auth, RLS, and storage. A key feature is the decoupled Node.js Voice Relay service, managing audio streams (e.g., Doubao ASR/TTS) and LLM interactions. This modularity supports flexible integration with various LLM providers (OpenAI compatible, Gemini, etc.), enhancing voice interview scalability.

Quick Start & Requirements

Docker Compose is recommended for self-hosting. Prerequisites include Node.js (>= 18), pnpm, and Docker Desktop. A configured Supabase project (URL, keys) and LLM API keys (e.g., OpenAI) are mandatory; Doubao keys are needed for voice. Initial Docker builds take 10-20 minutes. The GitHub repo is at https://github.com/yuanzhongqiao/ai-interview-platform.

Highlighted Details

• AI-driven interview generation from natural language descriptions.
• Multi-channel interviews: voice, text, Live Coding (Monaco), whiteboard (Excalidraw).
• Pluggable LLM architecture (OpenAI, Gemini, Kimi, MiniMax).
• REST API (OpenAPI spec) for ATS/HR integration.
• Anti-cheating features: screen-sharing detection, paste restrictions.

Maintenance & Community

Specific maintainer details, community channels (Discord/Slack), or sponsorships are not detailed in the provided README content. The repository is hosted on GitHub under yuanzhongqiao.

Licensing & Compatibility

Released under the MIT License, permitting commercial use, modification, and distribution, including integration within closed-source applications.

Limitations & Caveats

The README notes missing screenshots for key modules (Dashboard, Voice Interview, AI Analysis Report). Local development requires Supabase setup (local or cloud). Full functionality, especially voice interviews, depends on correct external API key configuration (e.g., Doubao).