A software engineer skilled in building CRUD RESTful APIs and is familiar with Swagger API documentation. Have hands-on experience in python web frameworks, Flask, Django and Bottle. Familiar with Linux development environment. Have good understanding of container(Docker). In addition, have solid knowledge of machine learning and Cloud DevOps.
Skills
Programming Languages
Python
90%
Go
70%
Java
60%
JavaScript
60%
C++
50%
Frameworks and Libraries
Tensorflow
80%
Keras
85%
PyTorch
70%
Scikit-learn
80%
Education
Certification of Computer Programming
Mar. 2018 – Dec. 2018
UCSC Silicon Valley Extension
Relevant Coursework:
Cloud DevOps Engineer Nanodegree
Sep. 2019 – Dec. 2019
Udacity
Relevant Coursework:
Machine Learning Nanodegree- Term 1
Jan. 2019 – Feb. 2019
Udacity
Relevant Coursework:
Deep Learning Nanodegree
Sep. 2018 – Jan. 2019
Udacity
Relevant Coursework:
AI Programming with Python Nanodegree
Apr. 2018 – Jul. 2018
Udacity
Relevant Coursework:
Experience
Software Engineer
Apr. 2019 – Mar. 2020
Gemini Data Inc., Taiwan
Material/Product Engineer
Jan. 2015 – Mar. 2018
Cymmetrik Enterprise Co., Ltd, San Jose, CA
Research and Development Engineer
Feb. 2011 – Aug. 2012
AUO, Hsinchu, Taiwan
Build a convolutional neural network(CNN) with Keras to classify dog breeds and then turn the code into a web app using Flask. Given an image of a dog, the model will identify an estimate of the canine’s breed. If supplied an image of a human, the code will identify the resembling dog breed and overlay filters with dog's ears, nose and tongue.
Build a neural network from scratch and use it to predict daily bike rental ridership.
The prediction fits great before Dec 22th. However, it shows obvious deviation from the data around the Christmas holiday. This model trained by the dataset may not have enough information about holidays to predict the effect of the holiday seasons.
Generate a new TV script for a scene at Moe's Tavern using Recurrent Neural Network(RNN).
Create a Generative Adversarial Networks(GANs) to generate new images of faces.
Generated images which can be obvious look like faces after running two epoches.
Develop a model to predict Boston housing prices. Utilize sklearn techniques for training, testing, evaluating and optimizing models.
Use Vgg19 and Xception pre-trained network to do artistic style transfer which obtain a representation of the style of an input image and apply the style to another image while keeping the original image content recognizable.
Identify the similarity among two thousands of documents based on the Word2Vec skip-gram model.
Evaluate 10 supervised learning algorithms that are currently available in scikit-learn on data collected for the U.S. census to help CharityML (a fictitious charity organization) identify people most likely to donate to their cause.
Implement unsupervised learning techniques on product spending data collected for customers of a wholesale distributor in Lisbon, Portugal to identify customer segments hidden in the data.