Welcome to my structured project ecosystem. All my work is gradually converging toward a modular, task-driven Agent Operating System (AOS) — a platform built for AI automation and software construction.

欢迎来到我的项目生态系统。目前我的所有工作正逐步汇聚为一个模块化、任务驱动的 智能代理操作系统 AOS（智核），它致力于 AI 自动化与软件构建。

A soft-kernel platform that unifies all my projects under a task-driven, multi-agent runtime.

一个软内核平台，以任务驱动和智能代理为核心，统一调度我所有的子项目。

AOS is the architectural foundation of my ecosystem. It is a soft-layer operating system for intelligent agent coordination, creation workflows, and system-wide orchestration — akin to what ROS is for robots, AOS is for AI-driven software and creative agents.

AOS 是我整个项目体系的架构核心，它提供智能代理的统一运行环境，调度任务与模块，形成面向意图的执行系统。就像 ROS 之于机器人，AOS 是创作型智能体的操作系统。

Core Responsibilities / 核心职责：

• Task management / 任务管理
• Agent lifecycle orchestration / Agent 生命周期调度
• Dependency-aware multi-module execution / 多模块依赖感知调度
• Execution feedback learning / 自我演化与行为修复机制

A system-level toolkit designed to extend AOS capabilities into the Linux kernel space.

一个用于扩展 AOS 到 Linux 内核侧的系统级智能工具包。

moeai-c is not the core engine but a specialized AgentPack under AOS. It focuses on OS-side automation, low-level task monitoring, and runtime feedback capture.

moeai-c 不再被视为“核心 AI 引擎”，而是 AOS 下的一个特殊 AgentPack，专注于操作系统自动化、内核交互与实时反馈。

• Kernel-level process scheduling augmentation / 进程调度增强
• Hooks for eBPF-based awareness (planned) / eBPF 感知（计划中）
• CLI tools for Linux automation / 命令行工具
• Seamless integration with AOS / 无缝接入 AOS 系统

An agent-based coordination platform for collaborative software engineering.

基于 Agent 的智能化软件构建系统，我硕士论文的核心平台。

A GitLab-driven, modular, resource-efficient multi-agent system designed for practical collaboration under limited compute environments.

基于 GitLab 的模块化 Agent 协作系统，适用于资源有限环境的实际测试。

DevNexusLite operates as an AgentPack within AOS. It simulates multi-agent orchestration using Git, Python, and minimal config.

DevNexusLite 在 AOS 中作为一个 AgentPack 运行，使用 Git + Python 模拟多代理协作。

graph TD
  subgraph 基础工具层
    E[Logloom<br>日志与多语言系统]
  end

  subgraph 核心引擎层
    A[moeai-c<br>内核智能工具集]
    B[KnowForge<br>知识结构引擎]
  end

  subgraph 应用与平台层
    F[LingCraft<br>AI学习系统]
    C[AIMADojo<br>AI自动剪辑器]
    D[DevNexus<br>软件工程代理系统]
    G[LingBridge<br>多语翻译系统]
  end

  subgraph 控制中枢
    Z[AOS<br>Agent Operating System<br>智能操作平台（软内核）]
  end

  D --> A
  D --> B
  D --> C
  D --> F
  D --> G

  F --> B
  F --> E
  C --> E
  A --> E
  B --> E
  D --> E
  G --> B
  G --> E

  Z --> A
  Z --> B
  Z --> C
  Z --> D
  Z --> E
  Z --> F
  Z --> G

A cross-language logging + multilingual interface layer

• Unified logging format across tools
• Designed for AI diagnostics, anomaly detection
• Supports Chinese/English/i18n message sets
• Version 1.0 released

AI-Powered Knowledge Refiner

• Convert PDFs, web, code into Markdown/Jupyter
• OCR + NLP + ChromaDB + LLM orchestration
• Integrated with moeai-c, used in DevNexus experiments
• Actively under development

Structured AI Learning System for Students & Language Learners

• Cross-platform intelligent notebook + adaptive learning engine
• Multimodal input: handwriting, screenshots, voice, structured text
• Built-in review scheduler and card-based practice flow
• Fully developed via DevNexusLite agents (GitLab-driven)
• Powered by KnowForge for OCR, semantic structuring, and concept mapping
• Notebook execution engine with Python + Pyodide
• Currently in early development

Bridging voices, beyond words. 连接声音，跨越语言。

LingBridge 是一个多平台智能翻译系统，聚焦于多语种语音/文字实时翻译、个体感知与上下文理解，旨在打破语言壁垒，构建沉浸式、自然流畅的沟通体验。未来将拓展至 VR、AR 与脑机接口等人机交互形态。

LingBridge is a cross-platform intelligent translation system designed for immersive, seamless multilingual communication. With real-time translation, contextual understanding, and speaker awareness, it bridges language gaps across physical and virtual worlds—including future integration with VR, AR, and neural interfaces.

Automatic MAD (Music Video) Editor

• Beat sync + cut detection + AI previewing
• Designed to be invoked from moeai-c
• Focused on anime/game music content
• Still in planning stage

CLI tool for gaming VMs under Linux

• Custom Windows VM builder
• For games with anti-cheat (e.g., Wuthering Waves, R6, APEX)
• Core scripts implemented, but outcome didn't meet expectations

• Advanced Machine Learning / 高级机器学习
• Linux Kernel Programming / Linux 内核编程
• Software Language Engineering / 软件语言工程
• Reinforcement Learning & Learning-based Control / 强化学习与基于学习的控制
• Concepts and Models for Parallel and Data-centric Programming / 并行与以数据为中心的编程模型

I expect to start my Master's thesis next semester. / 预计下季开始撰写硕士论文。

• 👯 Looking to collaborate on: AI agents, DSL tooling, OS-level automation
• 💼 Open to work in: AI, Game Development, VR, Parallel Programming, Systems Engineering
• 💬 Ask me about: Python automation, RL experiments, Linux dev, CS in Germany
• 📧 Email: ydzat@live.com
• 😄 Pronouns: he/him / 他
• ⚡ Fun fact: I once wrote a bot to snipe appointments from the German Ausländerbehörde... but they’re always on vacation. 😅

I'm passionate about games and aspire to work in the game industry in the future. I am currently prototyping a "fourth wall-breaking" narrative-driven game. Whether it will be released on Steam depends on the final development quality.

此外，我有多个关于交互叙事与玩法机制的游戏创意，目前尚待时间与资源支持。