Skip to content
View breath57's full-sized avatar

Block or report breath57

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don't include any personal information such as legal names or email addresses. Markdown supported. This note will be visible to only you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
breath57/README.md

👋 Hi, I'm breath57

Typing SVG

Email CSDN Visitors

💭 "In my view, truly great programmers can solve complex problems with limited technology. But this depends not only on technical depth and breadth, but on thinking about the essence of the problem. —— Breath57"

🧑‍💻 Tech Stack
class Breath57:
    def __init__(self):
        
        # Technical Breadth: AI Full-Stack
        self.tech_stack = {
            "backend": ["Go", "Python", "Java"],
            "frontend": ["Vue3", "WeChat Mini Program", "TypeScript"],
            "agent": ["LangChain", "LangGraph", "RAG", "Multi-Agent"],
            "ai": ["Pytorch", "Numpy", "Pandas", "Matpoltlib", "Sckit-learn", "Transformer"],
            "devops": ["Docker", "K8S", "GRPC", "RocketMQ"]
        }
        
        # Technical Depth: AI Agent Focus
        self.expertise = {
            "frameworks": ["LangChain", "LangGraph", "CrewAI", "Dify"],
            "focus": ["LLMOps Platform", "RAG Optimization", "Agent Workflow", "MCP Protocol"]
        }
        
        
        self.current_focus = "🎯 AI Agent - Building Intelligent & Efficient Agent Applications"
💼 Experience

Work Experience:

  • 🏢 Xiaomi - Golang Backend Engineer, Lead Tech of Reach Domain, supporting 10M+ level push services
  • 🏢 Yumc - Full-Stack Engineer (Java backend focused) for AI Intelligent Annotation Platform
🏆 Achievements
Honors
3rd Place in Xiaomi AI Agent Challenge
Chrome Plugin Promoted by Company's Level-1 Department, 38k+ Uses in 2.5 Months
Scholarship Every Year During School, Highest First-Class
Published Core Journal Paper on Deep Learning
📝 Tech Blogs & Thinking

"Not just writing code, but thinking about the essence of technology"

Featured Articles

🌟 Featured Personal Projects

Python library for video content retrieval and automatic PPT extraction

Innovation:

  • 💡 Proposed Bidirectional Text Similarity Algorithm
  • 🚀 Completed 1 year earlier than Baidu Netdisk's similar feature
  • 📊 Supports video subtitle matching and automatic PPT extraction

Tech: Python NLP OCR Video Processing

🔗 Documentation | Demo

Complete LLMOps platform inspired by Dify, demonstrating deep understanding of AI Agent tech stack

Core Features:

  • ✨ Custom FunctionCallAgent based on LangGraph
  • 🔧 Support for Plugins, Agent, Workflow, RAG, Streaming Chat
  • 🗃️ Dual Database (Weaviate + Postgres) for Hybrid Search
  • 🚀 One-click deployment with Docker Compose

Tech: LangChain LangGraph Flask Celery Vue3 Weaviate Docker

🔗 Live Demo | Repository

🚀 Fast Locate (In Development)

VS Code plugin for code organization across multiple service projects

Goals: Improve code reading efficiency | Quick cross-service navigation

💬 Let's Connect & Share Tech Insights!

📧 Email: breath57@163.com

Pinned Loading

  1. zereight/gitlab-mcp zereight/gitlab-mcp Public

    First gitlab mcp for you

    TypeScript 1.2k 230

  2. langgenius/dify langgenius/dify Public

    Production-ready platform for agentic workflow development.

    TypeScript 132k 20.5k

  3. langchain-ai/open_deep_research langchain-ai/open_deep_research Public

    Python 10.8k 1.6k

  4. langchain-ai/deepagents langchain-ai/deepagents Public

    Deep Agents is an agent harness built on langchain and langgraph. Deep Agents are equipped with a planning tool, a filesystem backend, and the ability to spawn subagents - making them well-equipped…

    Python 10.2k 1.6k

  5. dingtalk-skills dingtalk-skills Public

    🔥钉钉Skills库📚,含:文档、AI表格等,自动管理配置,轻量级—仅依赖curl, skill经过严格测试。

    Python 1

  6. how-ai-agents-remember how-ai-agents-remember Public

    源码级逆向工程(OpenClaw、nanobot、NullClaw、OpenFang)记忆系统 —— 架构图、数据模型、检索管线、复刻指南

    13