interview-guide
基于 Spring Boot 4.1、Java 25、Spring AI 2.0、React、PostgreSQL/pgvector、Redis 和 RustFS 构建的开源 AI 面试平台,支持简历智能分析、模拟面试、语音面试和知识库 R
What is interview-guide?
基于 Spring Boot 4.1、Java 25、Spring AI 2.0、React、PostgreSQL/pgvector、Redis 和 RustFS 构建的开源 AI 面试平台,支持简历智能分析、模拟面试、语音面试和知识库 RAG。
Overview of interview-guide
What it is
InterviewGuide is an open-source AI interview platform built with Spring Boot 4.1, Java 25, Spring AI 2.0, React, PostgreSQL/pgvector, Redis, and RustFS. It integrates resume analysis, mock interviews (text and voice), interview scheduling, knowledge base management, and RAG (Retrieval-Augmented Generation) for knowledge base Q&A. The system uses large language models (LLM), vector databases, Redis Stream for async tasks, and real-time voice technology to serve job seekers, HR, and training institutions.
Key features
- Resume management: supports PDF, DOCX, DOC, TXT formats; async processing via Redis Stream with progress tracking; automatic retry (up to 3 times) and duplicate detection based on content hash; PDF export of analysis reports.
- Mock interviews: Skill-driven question generation for 10+ directions (Java backend, Alibaba/ByteDance/Tencent specializations, Python, algorithms, system design, etc.); history question deduplication; configurable follow-up questions; unified evaluation engine for text and voice interviews; PDF report export.
- Voice interviews: real-time streaming dialogue via WebSocket with Qwen3 ASR/TTS/LLM; server-side VAD for automatic sentence segmentation; echo protection and manual submission; multi-turn context memory with pause/resume; Micrometer metrics for TTS/ASR latency.
- Interview scheduling: rule + AI dual-engine parsing of invitations (Feishu, Tencent Meeting, Zoom); calendar views (day/week/month) with drag-and-drop; status transitions with scheduled auto-expiry; configurable reminders.
- Knowledge base management: supports PDF, DOCX, Markdown upload, chunking, and async vectorization; RAG with pgvector using query rewriting, similarity threshold, and TopK; streaming responses via SSE; session management with pinning and multi-knowledge-base association.
- Knowledge base question bank and interview: generates questions from vectorized documents with main question, reference answer, key points, scoring criteria, and follow-ups; async generation with quality hints; strict capacity validation before interview; unified evaluation engine for scoring and feedback.
- Multi-model and system settings: built-in providers for DashScope, LM Studio, Kimi, DeepSeek, GLM; default model switching for chat and vector models; visual ASR/TTS configuration with connectivity test; configuration stored in
~/.interview-guide/with API key encryption.
Use cases
- Job seekers can upload resumes for AI analysis, practice with mock interviews (text or voice), and manage interview schedules.
- HR professionals can use the platform to evaluate candidates, schedule interviews, and generate assessment reports.
- Training institutions can build knowledge bases, create question banks, and conduct skill-specific mock interviews for students.
- Developers can study the project as a reference for building AI-powered applications with Spring Boot, Spring AI, and RAG.
Getting started
See the repository README for installation and quick-start instructions.
License and pricing
The project is licensed under AGPL-3.0. The README states that the full functionality is free and open-source, with no Pro version or paid core features. A paid tutorial is available for those who want systematic guidance.
Based on the project's public README and repository metadata.