ZJunCher

ZJunCher / xiaoyan-ai-dev-assistant

Public

基于 RAG 混合检索与多轮记忆的 AI 研发助手，支持团队知识问答，也适合新手学习 RAG 应用开发。

92
3
80% credibility
Found May 13, 2026 at 91 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Java
AI Summary

An AI chat assistant for development teams that uses uploaded documents and conversation memories to provide context-aware answers to project-related questions.

How It Works

1
💡 Discover XiaoYan AI Helper

You hear about this friendly AI sidekick that answers coding questions using your team's own guides and notes.

2
🖥️ Start the App

You download and launch the assistant on your computer, connecting it to your AI thinking service.

3
📤 Feed It Your Docs

You upload project manuals, training materials, and notes so it learns your team's ways.

4
💾 Save Personal Tips

You jot down favorite tricks or preferences as lasting memories for the AI to remember.

5
💬 Chat Away

You type natural questions about code, setups, or issues, and it pulls smart answers from your stuff.

✅ Boost Your Workflow

You and your team get quick, accurate help tailored to your projects, saving hours of searching.

Sign up to see the full architecture

4 more

Sign Up Free

Star Growth

See how this repo grew from 91 to 92 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is xiaoyan-ai-dev-assistant?

This Java Spring Boot app builds a RAG-powered AI dev assistant for team knowledge Q&A, pulling answers from uploaded docs via hybrid vector+BM25 search and multi-turn conversation memory. Upload files through the /api/documents endpoint, query via streaming /api/chat/stream, and add personal long-term memories at /api/memories—ideal as a rag github copilot alternative or rag github langchain example in Java. It runs local with in-memory vectors or scales to Pinecone, solving scattered dev docs and context loss in chats.

Why is it gaining traction?

Stands out as a rag github open source project with smart query rewrite for multi-intent questions, adaptive document chunking, and semantic dedup—delivering precise retrieval without the Python ecosystem lock-in of most rag github repos. Devs dig the ready-to-deploy API for team assistants, optional reranking, and fallback to local mode, making it a practical rag github local starter over bare LangChain4j setups.

Who should use this?

Backend Java devs building internal AI search for project docs, deployment guides, or troubleshooting. Team leads setting up shared knowledge bots to cut onboarding time. Newbies prototyping rag github project as a full-stack RAG assistant without gluing components.

Verdict

Grab it if you're in Java and need a battle-tested RAG base—92 stars show early promise, but tweak for production. Credibility score of 0.800000011920929% flags room for more tests/docs; solid learning tool, not enterprise-ready yet.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.