GV

GeezerVison

The World Wide Web, restored to 800 pixels of glory.

Home|About|Gaming cabinet|Archives|How it works|Starter sites

On this site

  • Sign in
  • ActionsCI/CD and automation
  • CodespacesCloud development environments
  • All Features
  • Documentation
  • Training
  • Certifications
  • Build with Copilot
  • Customize Copilot
  • Integrations
  • Trending repos
  • GitHub Sponsors
  • Developer articles
  • Enterprises

This view

  • Original page
  • New address

GeezerVison is showing github.com — chaduo - Overview

@chaduo chaduo Follow
More
  • Overview
  • Repositories
  • Projects
  • Packages
  • Stars
@chaduo
chaduo
Follow
View chaduo's full-sized avatar

Zhuo Zhao chaduo

Follow
My name is Zhuo Zhao . I am a master student majoring in microface at CUC in China.

Highlights

  • Pro
More
  • Overview
  • Repositories
  • Projects
  • Packages
  • Stars
chaduo/README.md

Hi there 👋 I'm Chaduo

🎯 AI Engineer | LLM Systems | Computer Vision

专注于大模型底层基座构建与智能体系统开发,关注:

  • LLM from scratch(Transformer / 推理优化)
  • Coding Agent / Tool-using Agent
  • 视频理解与微表情分析

🚀 Featured Projects

🔹 Mini Claude Code (Coding Agent)

  • 构建本地轻量级 Coding Agent(支持工具调用 + 长上下文)
  • 完成率从 72% → 88%
  • 上下文压缩降低 31%,高危操作拦截率 100%

🔹 Mini LLM Core

  • 从零实现 GPT-style Transformer(RoPE / RMSNorm / SwiGLU)
  • 完整训练 + 推理 pipeline
  • KV Cache 加速推理 2.5x(113 → 283 tokens/s)

🔹 Micro-Expression Detection

  • 双模态融合(VideoMAEv2 + 光流/应变)
  • Cross-Attention + Temporal FPN
  • 优化后处理提升定位性能

🧠 Tech Stack

🔧 Core

Python PyTorch

🤖 LLM & Systems

Transformer LLM Agent

⚡ Engineering

Linux Git


📊 GitHub Stats

GitHub stats

Top Langs


📌 About Me

  • 🔍 关注方向:LLM系统 / Agent / 多模态
  • ⚙️ 偏工程 + 算法结合(不是只调库)
  • 📈 持续优化推理效率与系统稳定性

📫 Contact

  • GitHub: https://github.com/chaduo

Pinned Loading

  1. CodePilot CodePilot Public

    Industrial-grade local AI coding agent with MCP, context compression, and permission control

    Python 1

  2. Mini-LLM-Core Mini-LLM-Core Public

    A from-scratch GPT-style LLM framework with tokenizer, Transformer, training pipeline, and KV-cache optimized inference.

    Python

  3. chaduo chaduo Public

Something went wrong, please refresh the page to try again.
If the problem persists, check the GitHub status page or contact support.

Uh oh!

There was an error while loading. Please reload this page.

You are visitor 01402503 — GeezerVison does not store the pages you look up.
Best viewed with Netscape Navigator 4.7 or Internet Explorer 6 at 800×600. No cookies required.