Snailclimb/interview-guide

What it solves

InterviewGuide is an intelligent AI interview platform designed to automate and enhance the recruitment and interview preparation process. It solves the difficulty of manually analyzing resumes, creating relevant interview questions, and conducting realistic mock interviews for candidates, HR professionals, and training institutions.

How it works

The system integrates Large Language Models (LLMs) with a RAG (Retrieval-Augmented Generation) architecture. It uses a Java/Spring Boot backend and a React frontend. Key technical mechanisms include:

• Resume Analysis: Uses asynchronous processing via Redis Streams to parse multiple document formats and generate AI-driven analysis reports.
• Mock Interviews: Employs "Skill-driven" question generation based on predefined domains (e.g., Java, AI Agents) and uses a unified evaluation engine to score both text and voice interviews.
• Voice Interaction: Implements real-time streaming dialogue using WebSockets and the Qwen3 voice model for ASR (Automatic Speech Recognition) and TTS (Text-to-Speech).
• Knowledge Base: Utilizes PostgreSQL with pgvector for storing and retrieving document embeddings, enabling the AI to answer questions based on specific uploaded knowledge sources.

Who it’s for

• Job Seekers: For uploading resumes to get feedback and practicing with AI-driven mock interviews.
• HR/Recruiters: For batch-analyzing resumes and evaluating candidate capabilities efficiently.
• Training Institutions: For managing specialized knowledge bases and providing structured interview training.

Highlights

• Multimodal Interviewing: Supports both text-based and real-time voice interviews with low-latency streaming.
• RAG-Powered Questioning: Generates interview questions and answers directly from a custom vectorized knowledge base.
• Skill-Driven Framework: Includes 10+ predefined interview directions with specific difficulty distributions and scopes.
• Comprehensive Tooling: Features an interview scheduler with AI-powered invitation parsing and PDF report exportation for analysis and evaluations.
• Flexible LLM Integration: Supports multiple providers including DashScope, DeepSeek, Kimi, and GLM via OpenAI-compatible configurations.