Chun-Liang Wu
650-***-**** • Stanford, California 94305 • Email: adjnmn@r.postjobfree.com
github.com/JoshWuuu • linkedin.com/in/Chun-Liang Wu EDUCATION
STANFORD UNIVERSITY Stanford, CA
MS, Civil and Environmental Engineering (GPA: 4/4.3) in Sustainable and Design Construction (Energy) Sep 2019 - Jun 2021 Coursework: Deep Learning, Machine Learning, Reinforcement Learning (In progress), Principles Data-Intensive System (In progress), Data Mining and Analysis, Design and Analysis of Algorithm, Renewable Energy, Client-Side Internet Technology NATIONAL CHIAO TUNG UNIVERSITY (NCTU), TAIWAN Taiwan B.S. in Civil Engineering (GPA: 3.9/4.0) Sep 2014 - Jun 2018 SKILLS
• Programming Skills: Python, R, SQL, C++, JavaScript, HTML, CSS, LaTeX, Swift
• Data Science Tools: Keras, Tensorflow, PyTorch, Scikit-learn, Tableau, Matplotlib, Pandas, A/B testing, Time-series analysis
• Database system skills: Spark, PostgreSQL,MySQL
EXPERIENCE
STANFORD UNIVERSITY Stanford, CA
Research Assistant - with Prof. Michael Lepech (Python - Tensorflow - Keras - Time-series analysis) Oct 2020 - Dec 2020
• Retrieved and manipulated massive unstructured data from different sources and formats (e.g. GEOJSON, JSON, CSV, XLSX)
• Applied LSTM and Autoregressive LSTM to predict the long-term air quality and traffic velocity and volume [Code]
• Achieved the accuracy of 95% R-Squared for air quality and 92% R-Squared for traffic velocity and volume
• Developed Mapbox-based platform to visualize the urban data (energy, housing price, weather, AQI, and traffic data) [Video] RISE ENGINEERING CONSULTANTS INC Taiwan
Data Analyst (Python - A/B testing - SQL) Sep 2018 - Dec 2018
• Conducted A/B testing to provide business insights of multiple designs of the construction project to the customers
• Maintained and connected the relationship of the construction cost database via MySQL
• Increased sales by 10% and decreased costs by 14% due to A/B testing and well-structured database NATIONAL TAIWAN UNIVERSITY Taiwan
Civil Engineering Computer-Aided Engineering (CECAE) Intern Jul 2017 - Aug 2017
• Utilized Dynamo visual programming to parametrically build the data-driven bridge model
• Managed and visualized data of the bridge structure with building information software, Tekla [Video]
• Streamline the construction workflow (save 10% of cost and 13% of time) through animation of the data-driven model PROJECTS
A RESNET-50-BASED TRANSFER LEARNING AND ENSEMBLE METHOD FOR STRUCTURAL IMAGES [CODE] Course Project - CS 230 Deep Learning (Python - Tensorflow - Keras) (Image Classification)
• Constructed baseline models (VGG-19, GoogleNet and ResNet-50) with transfer learning for classification
• Built ensemble model by stacking five best ResNet-50 models to enhance the performance of the classification
• Achieved 94% accuracy via ensemble model (better than the state of art by 1%) MACHINE LEARNING BASED CLASSIFICATION FOR SENTIMENTAL ANALYSIS OF IMDB REVIEWS [CODE] Course Project - CS 229 Machine Learning (Python - Scikit-learn) (NLP)
• Vectorized the text via binary, word-count and frequency-inverse document frequency (tf-idf) methods
• Utilized Logistic Regression, Support Vector Machine (SVM), Random Forest, Boosting, and Deep Neural Network(DNN)
• Achieved 91% accuracy through DNN with tf-idf (improvement of 1%) DATA ANALYSIS: A/B TESTING AND TIME SERIES ANALYSIS [CODE] Course Project - CEE 322 Data Analytics in Urban System (Python - A/B testing - Time-series analysis)
• Applied A/B testing to check the average treatment effect (ATE) of the water conservation program on two cities
• Utilized seasonal decomposition to observe the trend and random noise of wind and solar generation potential in two cities
• Implemented Auto correlation and Cross correlation to check the predictability of wind and solar generation A REAL-TIME SMART GRID SYSTEMVIAARTIFICIALINTELLIGENCEINTAIWAN Course Project - CEE 277 Smart Cities and Communities
• Implemented DNN to predict future hourly energy generation and energy demand with 90 % R-Sqaured
• Applied Deep Reinforcement Learning to optimize the operating schedule for the appliances (save 20% of electricity cost)