An AI application developer and explorer focused on industrial landing of artificial intelligence. Currently, my core research and practice focus on LLM application implementation, RAG question-answering, multi-agent systems, and lightweight model inference deployment, with rich practical experience in NLP, OCR, document intelligence and knowledge graph.
- 🤖 LLM / RAG / Multi-Agent Systems (LangChain & LangGraph)
- 📝 Natural Language Processing & Text Intelligence
- 🖼️ Document Image Processing & Computer Vision
- 📊 Data Mining, Search Engine & Personalized Recommendation
- 🧩 Knowledge Graph Construction & Intelligent Q&A
- 📚 GitHub : jiangnanboy.github.io
- 🌰 My Website: wexopen.com
- 📘 Zhihu ：https://www.zhihu.com/people/jiangnanboy
- 📘 CSDN ：https://blog.csdn.net/qq_20182781
| Project Name | Introduction | Online Address |
|---|---|---|
| TalkSheet Agent | Excel Intelligent Analysis Agent | https://www.talksheet.wexopen.com |
| RAG QA System Agent | Industrial Intelligent Q&A Agent | https://www.rag.wexopen.com |
| Multi-Agent Office System | Office collaborative intelligent agent platform | https://www.office.wexopen.com |
| Project | Description |
|---|---|
| OCR & Table Tool | All-in-one OCR, table recognition, screenshot capture & export |
| Video to PPT/PDF Converter | Generate editable PPT from video with deduplication & clip extraction |
| Document Comparison Tool | Multi-format document contrast analysis & intelligent report generation |
| Document Image Processor | Image enhancement, distortion correction, shadow removal & edge cropping |
| Project | Description |
|---|---|
| RAGForge | Modular, composable RAG retrieval library —— Build your retrieval pipeline like forging parts |
| Chinese Text Corrector | Practical open-source Chinese error correction toolkit |
| Intelligent Medical KG | Medical knowledge graph, symptom retrieval and diagnosis Q&A |
| K12 Education Knowledge Graph | Education-oriented KG visualization and knowledge reasoning |
| OpenOCR4j | Lightweight end-to-end document parsing |
| JiaJiaOCR | Pure Java self-developed OCR recognition engine |
| Document Image Tool | Scanned document enhancement and batch optimization |
| Learning to Rank | Search sorting model optimization and algorithm practice |
| Project | Description |
|---|---|
| All projects | Project Classification Details |
- 📧 Email: 2229029156@qq.com
- 💬 WeChat Official Account：番石榴AI
- 🐙 GitHub: https://github.com/jiangnanboy