about me
HI! I AM HANK
Permanent resident & work authorization was approved on July
Permanent resident & work authorization was approved on July
Hi, I'm Hank, an experienced research assistant with a demonstrated history of working in the higher education industry. Comfortable programming in Java and Python. Skilled in Full Stack Web Development, System Design, Machine Learning and Computer Vision. Technology used include HTML5, CSS3, JavaScript, Node.js, Express.js, REST API, SQL/noSQL Database, BootStrap4. Received a Master's Degree in Computer Science with GPA 4.0.
Project Experience: Full Stack Web Development, System Design, Artificial Intelligence, Machine Learning
Certificate: AWS Certified Cloud Practitioner (2021), The Complete Web Development Bootcamp (2020)
Software Engineering
Web Development
Artificial Intelligence
Machine Learning
System Design
Computer Vision
what I do?
my TECHNICAL skills
JAVA
- 10000+ lines of code(loc)
- 150+ Leetcode problems
- Largest Progect: 6000+ loc
- Maven projects
- Artificial intelligence projects
Python
- 10000+ lines of code(loc)
- 50+ Leetcode problems
- Largest Progect: 3000+ loc
- Computer vision and machine learning projects
C++
- 4 years undergraduate program code in C++
- 3D computer graphic code with OpenGL and GLSL for shading, rendering and ray tracing.
Web Development
- Font-end: HTML, CSS, JS, React, Angular.js
- Backend: AWS, nodejs, SQL, MongoDB
- Attended Caltech and USC hackathon
Machine Learing
- Library: Keras, Tensorflow
- Tools: numpy, Pandas, scikit-learn, scikit-image
- Top 3 on the Kaggle class competitions
- One Research Poster ➜
Computer Vision
- Image and video processing
- Image segmentation, Image matting and Object tracking
- 3D Computer Graphic
- One Research Poster ➜
previous work
WORK EXPERIENCE
(2020-2021 applying for permanent resident card)
(2020-2021 applying for permanent resident card)
Freelance Software Engineer
Jan 2020 - Oct 2021
Freelance, Glendora, CA
- Adopted test-driven development and built interactive website by using React, Node.js, Expreess.js, Bootstrap and MongoDB.
- Obtained AWS Certified Cloud Practitioner certification. Completed Udemy course for Web Development Bootcamp.
Research Assistant
Dec 2017 - Dec 2019
California State Polytechnic University, Pomona, CA
- Studied in Computational Intelligence Lab adviced by Prof. Hao Ji.
- Improve object segmentation based on FCN and image matting. Increase 2% accuracy compare to Mask R-CNN and SiamMask.
- Develop an auto data preparation method for machine learning object detection by creating an automated method leveraging 3D model reconstruction and computer graphics to synthesize images for any custom objects.
- Implement real-time object tracking method based on computer vision and fully convolutional Siamese networks.
- Create an object detector for an industry project to separate coals and rocks in images by applying image filtering, image threshold, and image transformation.
Teaching Assistant & Grader
May 2018 - Dec 2019
California State Polytechnic University, Pomona, CA
- Prepared 38 data structure and algorithm assignments for Python programming.
- Assist professor to review quizzes in the class and grade tests, assignments and projects for a variety of courses as below:
○ CS 4990 – Machine Learning (Python, Keras)
○ CS 2400 – Data Structures and Advanced Programming (JAVA)
○ CS 264 – Computer Organization and Assembly Programming (Mips)
○ CS 241 – Data Structures and Algorithms (JAVA)
Information and Technology Engineer
Feb 2015 - Mar 2015
World Vision, Taipei, Taiwan
- Collected event data and online information also managed personal registration records.
- Maintained computers and network, including resolving equipment malfunctions, providing media access, and corresponded with multiple users to transmit and process information.
my work
here's what i've done
Full Stack Technology Profile
Last Update: November 16, 2021
- This web application used Node.js and Express.js to build REST API as backend interacts with MongoDB Atlas and Mailchimp API.
- This project also applied session and data encryption for information security. Frontend technology includes HTML5, CSS3 and JavaScript for responsive website and animation objects.
Machine Learning: Segmentation & Classification
Published: May 28, 2019
- Face & Hair Segmentation: Built a U-Net and VGG model using Keras and applied callback functions to monitor the training procedure. Increase the accuracy 20% by freezing the intermediate layers and using data augmentation and facial landmark detection to provide more information to the training model.
- Gender Classification: Built a VGG-16 model with a pre-trained weight using Keras and improved accuracy 2% through applied face detection to remove 5% none relate images from the dataset.
Artificial Intelligence Game Strategy (3 Games)
Published: Jan 31, 2019
- Four-In-A-Line: Used minimax algorithm, customized evaluation function to generate a decision tree, and applied alpha-beta pruning to reduce the running time. The first-hand AI achieved a win rate of 60%.
- N-Queen: Applied genetic algorithm solved N-queen problem then reduced the running time 97% by using simulated annealing.
- 8-puzzle: Used 'A' star algorithm with 2 different heuristic functions to successfully solve an 8-puzzle problem.
Scheduling Application
Published: Mar 14, 2018
- The team project for the system is designed to create a meeting scheduler with graphical user interface for the company. My role was team lead and lead developer, mainly in charge of scheduling algorithm design and system development.
- Users can create meetings and invite other employees. The invited employee can choose to either accept, decline, or respond to the invitation. The system will assign a room depend on the time and the capacity. [Java, Maven, MongoDB]
my research
computational intelligence Laboratory
MatRCNN: Moving Object Removal in Video Sequence
Presented on Aug 8, 2019
Abstruct
MatRCNN is shot for Image Matting and Regional Convolutional Neural Network. Removing undesired objects from video sequence plays an important role in many real-world applications. However, many existing methods require users to provide mask information annotating object areas in video frames, which appear to be extremely time-consuming and tedious. The purpose of this study is to develop an automatic framework that removes moving objects from a video with minimal human intervention, by taking advantage of the power of deep learning. Image matting and object tracking are also investigated to improve the quality of the object masks generated by deep models and to eliminate artifacts in the resulting videos.
Automated Data Preparation for Custom Object Detection
Presented on Aug 16, 2018
Abstruct
Training a custom object detector requires a large dataset of images of the objects and their corresponding bounding box annotations. However, annotating images with bounding boxes around objects which is currently performed manually, appears to be very time-consuming and tedious. We propose an automated method leveraging 3D model reconstruction and computer graphics to synthesize images and their bounding boxes for any custom objects. Our experimental results show that the proposed automated data preparation achieved compelling detection accuracy with limited user assistance.
Academic achievement
education & awards
Master Degree of Computer Science
California State Polytechnic University, Pomona
GPA: 4.0/4.0
Bachelor Degree of Computer Science and Engineering
National Taiwan Ocean University
GPA: 3.6/4.0
Awards & Certificate
Academic Excellence Award, 2014
Ranking: 2/54
AWS Certified Cloud Practitioner, 2021
Amazon Cloud Service
extracurricular activities
activity & society
Event Coordinator
2018 - 2019
International Christian Fellowship ( CPP )
• Communicated with campus and other organization, helped the club register the event and decide the location.
Student Representative
2014 - 2015
Department Student Association ( NTOU )
• Provided assistance and counseling for student issues within the department. Acted as mediator between students and university.
Volunteer & Team Follower
March 2015
Int. Ice Hockey Federation
• Helped participating countries and assisted teams with check-in, reviewed shcedules and communicated between team and organizer.
Director
May 2013
Computer Science Night ( NTOU )
• In caharged of scenario editing, lighting control and music design, also direted a drama for the event. 70 students participated and 300 audiences .
Organizer
March 2013
One Day University Camp ( NTOU )
• Designed and planned a camp for 30 high school students to participate in the "Experience the University Life and University Tour".
Financial Officer
Sep 2012 - Sep 2013
High School Alumni Association ( NTOU )
• Management of association funding and budget review, and regularly publishing income and expenditures.
◉
get in touch
contact me
Yueh-Lin Tsou (Hank)
hankytsou@gmail.com
(415)-619-8030
Copyright © 2019-2021 All rights reserved | This Website is made by Hank Tsou