DATA SCIENTIST

MARWIN KO

Experience

Metis San Francisco, CA

Data Scientist Jul 2016 to Sep 2016

Metis is a 12-week immersive data science fellowship focusing on machine learning, statistical modeling and data acquisition through Python. Developed ve data science projects using skills such as machine learning, natural language processing, and web scraping. MESA Laboratory Merced, CA

Research Statistician Aug 2013 to Dec 2015

Working in the Mechatronics, Embedded Systems, & Automation (MESA) Laboratory, conducted research that involved signal processing techniques applied to physiological signals, such as heart rate variability

(HRV).

Cleaned, processed, and applied fractional-ordered algorithms to HRV time series in MATLAB. Classi ed and differentiated between human subjects with or without cardiac arrhythmia. Published thesis titled: Applications of Long Range Dependence Characterization in Thermal Imaging & Heart Rate variability.

UC Merced Bioengineering Laboratory Merced, CA

Research Data Analyst Nov 2011 to Aug 2013

Worked as a researcher and assisted in cell culturing, dynamic light scattering (DLS), and data analysis on various projects.

Gathered, cleaned, and presented data to other researchers. Used DLS to measure aggregation of polymers such as exopolymeric substances (EPS) and mucin. Resulted in identifying speci c chemical compounds that promoted polymeric aggregation. COINS Summer Internship Merced, CA

Research Engineer Intern May 2012 to Aug 2012

Participated in the Center of Integrated Nanomechanical Systems (COINS) summer internship, hosted by UC Berkeley, UC Merced, and Caltech.

Using electrospinning techniques, engineered a multitude of nano- brous stem cell scaffolds. These scaffolds were able to help promote stem cell proliferation and differentiation. Empirical work published

Projects

Tanzanian Ministry of Water Water Pump Classification (current project) This is a competition hosted by DataDriven. Using data from the Tanzanian Ministry of Water, I am building a model to predict which water pumps are function, need repair, or broken. Walmart Product Classification Using Text Description This was a global competition hosted by Walmart Labs via Hacker Rank with the objective to predict product labels. Utilizing product text descriptions and natural language processing (NLP) I trained a model to classify product labels. I placed in the top 100. AllState Insurance Claims Severity Loss Prediction This was a Kaggle competition hosted by AllState. I ran an exploratory data analysis (EDA) and created a machine learning model to predict insurance claim . Aerial Intelligence Wheat Yield Prediction Using Satellite Data Aerial Intelligence posted satellite data on Github with the challenge to predict wheat yield in the winter. I reduced the features using algorithms and trained several regression models using Random Forest and Gradient Boosting.

Metis NFL Draft Prediction

Built several regression models using college football and Nation Football League (NFL) data to predict draft pick of a player.

Summary

Data Scientist with four years of research

experience in bioengineering and mechanical

engineering. Love getting dirty with data and

using machine learning to help solve complex

business problems and provide actionable insight.

Contact

San Francisco, CA

acydfw@r.postjobfree.com

510-***-****

marwin_ko

marwinko

marwin-ko/projects

Education

University of California, Merced

BS Biological Engineering 2013

MS Mechanical Engineering 2015

Skills

PROGRAMMING

Python

Git

MATLAB

R (learning)

Linux Command Line

DATA DISTRIBUTION

Amazon Web Services (AWS)

NoSQL

MySQL

Spark (learning)

DATA SCIENCE

Machine Learning

Natural Language Processing

Web Scraping

CERTIFICATIONS

Oracle mySQL Fundamentals (in progress)

AWS Solutions Architect (in progress)

Contact this candidate