Biography

My research topic is data effinciency in spatio-temporal domain focusing on computervision

I am a second year PhD student in the WiNS Lab, the University of North Carolina at Charlotte, advised by Prof. Pu Wang. Prior to UNCC, I spent many wonderful years in Cox Automotive as a Machine Learning Engineer

EDUCATION

University of North Carolina at Charlotte, Charlotte, North Carolina, January 2021 - Present
Ph.D. in Computing and Information Systems

Stevens Institute of Technology, Hoboken, New Jersey, January 2016 - May 2017
Master of Science in Software Engineering

Sirindhorn International Institute of Technology, Bangkok, Thailand, June 2004 - May 2008
Bachelor of Engineer in Communication Engineering

WORK EXPERIENCE

Cox Automotive, New Hyde Park, New York
Machine Learning Engineer & Senior Full Stack Software Engineer, 09/2017 - 11/2020
  • Created machine learning and statistical modeling to predict probability of booking contracts and deploy machine learning model on AWS using Sage maker, Lambda, API gateway, and EC2
  • Developed a machine learning model to classify vehicle images and auto-fill the form for customers using Pytorch with fast.ai techniques such as auto find learning rate, dynamic learning rate, different learning rate for each layer, freeze layers, precompute activation, and test time augmentation running on AWS P2 instance
  • Designed internal React framework architecture to be able to consume and communicate between a separated self-contain application on private npm supporting Redux and Redux-saga
  • Refactored the profitable legacy application by decoupling data pipeline and the view components by running a new architecture in parallel without server downtime
  • Guided engineers to follow team design pattern and company architecture
  • Implemented continuous integration using Jenkins for automating the deployment
  • Developed Digital Showroom web application to facilitate customers to finish all paperwork online
  • Stevens Institute of Technology, Hoboken, New Jersey
    Research Assistant, 01/2016 - 05/2017
    Research Interests: Reinforcement Learning and Computer Vision
  • Researched on deep reinforcement learning focusing on minimizing agent experiences method
  • Implemented Face Recognition and Object Detection algorithm for machine vision using FaceNet and YOLO
  • eveloped centralized robotic control platform to support physical robots with video streaming using WebRTC, Opencv, ROS, Turtlebot, Raspberry Pi, and Gazebo Simulation
  • Area80 Co., Ltd, Bangkok, Thailand
    Senior Full Stack Software Engineer, 04/2011 – 02/2015
  • Designed software architecture using MVC, Universal Javascript, Server Rendering (Isomorphic), Load Balancing, Cluster, Static Cache Files, TDD, and DevOps
  • Developed front-end, back-end websites and embedded system using HTML, Javascript, AS3, PHP, MySql, Node.js, MongoDB, and Arduino such as CMSs, websites, game applications, motion detection, augmented reality, kinect, Wii gun, face detection, and sound recording
  • Created mobile game application using Adobe Air for Android and iOS
  • Designed back-end architecture using Codeigniter for PHP and designed company’s MVC front-end framework to offer the best user experience (UX)
  • Design Sundae Co., Ltd, Bangkok, Thailand
    Web Developer, 12/2010 – 03/2011
  • Developed interactive websites and games using AS3, HTML, Javascript, and PHP
  • Created animation using Adobe Flash, Illustrator, and Photoshop
  • Global Wireless Co., Ltd, Bangkok, Thailand
    Front-end Leader, 06/2008–11/2010
  • Developed e-learning framework support real-time video streaming, interactive games and exams
  • Managed work-flow and program structure of animation and programming in Agile team
  • Progaming Co., Ltd, Bangkok, Thailand
    Instructor, 06/2008–11/2010
  • Taught “Fundamental Action Script 3” course
  • Github

    Neural Network

    Computer Vision

    Math

    Contact Me

    9201 University City Blvd, Charlotte, NC 28223

    [email protected]

    © Ekkasit Pinyoanuntapong. All Rights Reserved.

    Designed by HTML Codex