Pham Van Doanh

AI Engineer at FPT Corporation
  • Claim this Profile
Online Presence
Contact Information
us****@****om
(386) 825-5501
Location
VN

Topline Score

Bio

Generated by
Topline AI

0

/5.0
/ Based on 0 ratings
  • (0)
  • (0)
  • (0)
  • (0)
  • (0)

Filter reviews by:

No reviews to display There are currently no reviews available.

0

/5.0
/ Based on 0 ratings
  • (0)
  • (0)
  • (0)
  • (0)
  • (0)

Filter reviews by:

No reviews to display There are currently no reviews available.
You need to have a working account to view this content. Click here to join now

Credentials

  • Deploying Machine Learning Models in Production
    Coursera
    Apr, 2023
    - Sep, 2024
  • Machine Learning Data Lifecycle in Production
    Coursera
    Mar, 2023
    - Sep, 2024
  • Project Management Essentials
    Udemy
    Jun, 2022
    - Sep, 2024
  • Introduction to Machine Learning in Production
    Coursera
    Sep, 2021
    - Sep, 2024
  • Convolutional Neural Networks (CNN)
    Coursera
    Apr, 2021
    - Sep, 2024
  • Structuring Machine Learning Projects
    Coursera
    Apr, 2021
    - Sep, 2024
  • Data Visualization
    Kaggle
    Feb, 2021
    - Sep, 2024
  • Intro to Deep Learning
    Kaggle
    Dec, 2020
    - Sep, 2024

Experience

  • FPT Corporation
    • Vietnam
    • Information Technology & Services
    • 200 - 300 Employee
    • AI Engineer
      • Apr 2022 - Present

      An AI system for FPT Telecom's fiber-optic box can recognize damaged details inside thebox through images. The system is applied to 700,000 boxes nationwide, supporting FPT'sinternet to always be continuous and uninterrupted. The system achieved an impressiveoverall accuracy of 93.5%.Tasks:- Analyze requirements and design system architecture.- Definition and instruction for labeling each of the 151 classes.- Clean the data, augment the data, and handle data imbalance.- Train and optimize the model.- Post-processing to recognize the details of the damage.- Deploy Microservices with Docker.Technology: Object detection, Image classification, Deep Learning, Pytorch, Python, Docker,Microservice API. Show less

    • AI Engineer
      • May 2021 - Apr 2022

      A video understanding system is being built to process and recognize context from FPT Playvideos. The system has a performance precision of 98%.Tasks:- Extract frames from FPT Play videos.- Pre-process CLIP with Multithreading.- Building a system to recognize actions, objects, and places.- Post-processing generates the start and end times of each context.Technology: Action recognition, Object detection, Place detection, Deep Learning, 3D CNN,Pytorch, TensorRT, Python. Show less

  • AIOZ AI
    • Singapore
    • Technology, Information and Media
    • 1 - 100 Employee
    • Robotics & AI Engineer
      • May 2019 - Mar 2021

      Working on some Robotics, Computer Vision, and Deep Learning projects: - Image processing. - Object classification. - Object detection. - Object segmentation. - Robot simulation. Working on some Robotics, Computer Vision, and Deep Learning projects: - Image processing. - Object classification. - Object detection. - Object segmentation. - Robot simulation.

  • National Chung Cheng University
    • Taiwan
    • Education Management
    • 100 - 200 Employee
    • AI Intern
      • Aug 2018 - Dec 2018

      Full scholarship by Taiwan Government for researching internship at ACM lab focused on Machine Learning - Deep Learning - Computer Vision. Project: THE WAITER ROBOT ARM. Description: Detects the QR code location in the space, the robot arm automatically takes the object and run to the QR code. Tasks: - Detect the positions of QRcode in 2D image. - Find the Intrinsic and Extrinsic of the camera using the Calibration camera method and Forward Kinematic method to define the QRcode positions in Robot coordinates. Technology: Calibration, Homography, Kinematic, C++ Show less

  • HCMC University of Technology and Education
    • Vietnam
    • Higher Education
    • Thesis: Parking lot management using Image Processing
      • Jan 2018 - Jun 2018

      Description: An image processing system that detects the vacant positions in the parking lot and returns the results to users on the Desktop interface. Tasks: - Image processing: process the images returned from cameras. Convert them into gray images and reduce the interference. - Cut the parking lot part from these images to figure out the change of gray-level in this part. - Set a value to notify whether cars are parking at that place or not by using this change of gray-level. Technology: Noise reduction, Segmentation, Feature extraction, Classification, Python. Show less

Education

  • HCMC University of Technology and Education
    Bachelor of Engineering - BE, Computer Engineering Technologies/Technicians
    2014 - 2018

Community

You need to have a working account to view this content. Click here to join now