Skip to content
View xiang-wuu's full-sized avatar
💭
I may be slow to respond.
💭
I may be slow to respond.
  • Conneqt Digital (QUESS CORP)

Block or report xiang-wuu

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.

Please don't include any personal information such as legal names or email addresses. Maximum 100 characters, 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
xiang-wuu/README.md

Hi there 👋

I'm Engineer & an AI Enthusiast with interest in applied AI and software development. I love working with and contributing to open source projects. Professional Computer Vision & ML/DL Expert with years of experience in researching and implementing optimized deployment solutions for various target platforms. Proven experience in deploying solutions to NVIDIA TensorRT, Intel Openvino, Android NDK's.

class Xiang(Minimalist):
  def __init__(self):
    self.professional_status = {
      "role" : "Sr. Digital Engineer", 
      "company" : "Conneqt Digital",
    }
    self.degree_status = {
      "level" : "BE", 
      "subject" : "Computer Science",
      "institution" : "SGB University",
    }
    self.languages = ["python", "java", "C/C++"]
    self.py_datasci_stack = ["opencv", "numpy", "scikit-learn", "keras", "pytorch"]
    self.tools = ["dvc", "git", "android-studio", "vs-code"]
    
  def main(self):
    if not self:
      self = self.__init__()

My Specializations & Expertise

  • Understanding the problem statement & finding best optimal solution.
  • Technical decision making & finalizing solution technology.
  • Leading ML engineers along with MLOps data associates.
  • Evaluating and Researching with various ML/DL algorithms which suits high level problem requirements.
  • Responsible for establishing and managing MLOps and complex ML workflow pipeline along with source version control system.
  • Leading team of ML Developers & Feature Engineers along with manual data annotation developers.
  • Experience in working with PyTorch deep learning framework for design and implementation of CNN's for image recognition, classification, retrieval problems.
  • Deciding optimization strategies for complex computer vision based deep learning problems.
  • Experience in deploying pre-trained NN models to highly optimized inference engines like NVIDIA TensorRT, Intel Openvino, TFlite and Android NDK's.

⚡ Some of my notable open source contributions:

  • Tensorrtx(4.6k Stars) - Integrated new Yolov5-v6.2 classification model support to infer using TensorRT IE (PR). SuperPoint Pytorch implementation using TensorRT IE (PR)

  • Tencent/Ncnn(15.5k Stars) - Added sample android based application which infers Yolov7 PyTorch model. (PR), the android sample is hosted at xiang-wuu/ncnn-android-yolov7

  • Yolov5 (30k Stars) - Added Gardio integration with hugging face spaces to offecial Yolov5 implementation.(PR)

  • Yolov6 (4k Stars) - Added DVC integration support to track data/model artifacts & support for MLOps integration. (PR)

  • Yolov7 (5.5k Stars) - Added TensorRT inference support to offecial Yolov7 implementation. (PR), also added some very basic functionality (PR)

  • xiang-wuu/TensorRT-PTQ🔒 - Post Training Quantization inference using TensorRT for different CNN Architectures.

  • xiang-wuu/opencv-projects🔒 - Collection of many usefull computer vision based projects for study and academic purposes.

  • xiang-wuu/activity-monitor-cpp🔒 - This is CPP implementation of User activity monitoring utility based on facial recogntion, this is primarily used to keep track of AFK/AFD user activity.

Private Work Portfolio:

⚡:1st_place_medal: Deployment Releases

⚡:1st_place_medal: Research Releases

Skills

python cplusplus Java mysql aws Shell CLion android-studio pycharm cmake pytorch tensorflow keras numpy opencv codecov git github github actions html Windows linux ubuntu

Stats

Socials

@xiang-wuu's activity is private