Skip to content
View Puiching-Memory's full-sized avatar
🎯
Focusing
🎯
Focusing

Highlights

  • Pro

Block or report Puiching-Memory

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
Puiching-Memory/README.md

Hello, 你好! 👋 这里是 Puiching-Memory

@梦归云帆

Visitor Count


GitHub Stats Top Languages



Profile Details
Repos per Language Most Commit Language

🛠 Tech Stack

Languages & Core: Python, C++, Cython
AI & Deep Learning: PyTorch, LLM/SD Fine-tuning, Agent Workflow, YOLO, VLM
Dev & Ops: Docker, Git, ECS, Electron, FastAPI, QT

🎓 Education

Guangdong Polytechnic Normal University (广东技术师范大学)
B.E. in Network Engineering (网络工程) | 2023 - 2027
Qualifications: CET-4 | Driving License C2

💼 Experience

  • R&D Intern @ Desay SV (德赛西威 002920.SZ) 2024.07 - 2024.09
    • Developed traffic sign detection system based on YOLOv5, extended for multi-label classification (40+ classes).
    • Built End-to-end Driving Agent using VLM with LoRA fine-tuning and automated data pipelines.
    • Implemented 2D/3D Monocular Lane Detection models and MonoDLE based 3D Object Detection.

🚀 Selected Projects

  • MonoLite 2024.09 - Present

    • Monocular 3D Object Detection System with modular architecture.
    • Implemented custom OPT caching and Cython acceleration for performance optimization.
    • Open Source: GitHub
  • PRfire (Smart Agriculture System) 2024.10 - 2025.03

    • Backend development using FastAPI & Docker, integrated VLM & YOLO for disease detection.
    • 3D Terrain Reconstruction using COLMAP + OpenMVS.
    • Open Source: GitHub
  • FakeBrain (Unity Game Backend) 2025.03 - 2025.05

    • Brain-Computer Interface backend using C++ & Qt.
    • LSTM based blink detection deployed via ONNX Runtime.
    • Game agent training pipeline with VLM.
    • Open Source: GitHub

🏆 Honors & Awards

Award Rank/Level Year
Tencent Game Algo Competition (Text2SQL) 54th 2025
"Challenge Cup" Competition (AI+ Track) 33rd 2025
Chinese Collegiate Computer Design Contest Provincial 3rd Prize 2023
National Undergraduate Electronics Design Contest Provincial 3rd Prize 2025
Lanqiao Cup (Python Track) Provincial 3rd Prize 2023

🌟 Community

SwanLab Community Contributor
Active Contributor [ 4 commits | 173++ | 142-- | #10 ]

📄 Publications

  • Zhang Kai, et al. (2025). Research on Narrowband Satellite IoT Video Transmission Technology Based on an Optimized AVS3 Model. IEEE IoTAAI Conference. (4th Author)


Last edit in 2026.02.08

Pinned Loading

  1. OPT4TorchDataset OPT4TorchDataset Public

    Plug-and-Play Optimal Page Replacement Algorithm (OPT) for Torch Dataset

    Python

  2. pylibde265 pylibde265 Public

    Decode HEVC(H.265) video in python \\ 在python中解码HEVC(H.265)视频

    C 5

  3. monolite monolite Public

    Explore lightweight practices for monocular 3D detection \\ 探索单目3D检测的轻量级实践

    Python 3 1

  4. SurviveGPNUManual SurviveGPNUManual Public

    广师大生存手册 \\ SurviveGPNUManual

    TypeScript 4 2

  5. TGAC_2025 TGAC_2025 Public

    TGAC2025腾讯游戏算法大赛 数智决策科学赛道 参赛队伍

    Python 2

  6. Tencent_Recommendation_2025 Tencent_Recommendation_2025 Public archive

    2025腾讯广告算法大赛_参赛队伍之一

    Python 8 1