John Sheng
Machine Learning Engineer at Tapad- Claim this Profile
Click to upgrade to our gold package
for the full feature experience.
Topline Score
Bio
Credentials
-
TensorFlow in Practice Specialization
CourseraAug, 2019- Nov, 2024 -
Deep Learning Specialization
CourseraJul, 2019- Nov, 2024
Experience
-
Tapad
-
United States
-
Advertising Services
-
1 - 100 Employee
-
Machine Learning Engineer
-
Oct 2021 - Present
-
-
-
Analytic Partners
-
Advertising Services
-
300 - 400 Employee
-
Data Scientist/Machine Learning Engineer
-
May 2020 - Oct 2021
- Built marketing mix product both backend and frontend for business team to report ROI to clients. - Build general machine learning platform both backend and frontend for business team. - Built sentimental analysis product backend for business team to monitor our clients’ products. - Built chatbot backend integrate with marketing mix and general machine learning platform. - Built marketing mix product both backend and frontend for business team to report ROI to clients. - Build general machine learning platform both backend and frontend for business team. - Built sentimental analysis product backend for business team to monitor our clients’ products. - Built chatbot backend integrate with marketing mix and general machine learning platform.
-
-
-
UTOFUN
-
United States
-
Real Estate
-
1 - 100 Employee
-
Data Scientist Intern
-
May 2019 - Aug 2019
Daily responsibility: data integrity by excel or python, data extract by SQL, data visualization repot by tableau. Project: Based on customer search behaviors on company website, analyze which feature mostly influence customer decision that they will finally contact agent. Result: based on analysis results to let engineer change web design style. Finally, improved 10% monthly company website searching volume and also improved contact agents volume. Method: Extracted data and merge data in SQL. Basically, clean and EDA. Building Logistic Regression, Random Forest and XGB model to analysis feature importance. Show less
-
-
-
Francis Peltast Partners
-
Greater New York City Area
-
Data Analytics Intern
-
Oct 2018 - Dec 2018
Built a regression model to predict the enrollment trend of undergraduate students in USA. First, selected predictor among number of male and female, students status (full or part-time), citizenship (international or not), number of higher GPA students (above 3.5). Second, based on stepwise process to add predictor into model, using t-test for stop rule. Third, calculated 2018-2020 predicted number and plot the linear regression model properly. Web scraping six different arears real estate information like price, number of bedroom, broker, etc using R. Analyzed results based on average of price/square, price/bedroom, and price/bathroom and plot three properly histograms based on ggplot package in R. Show less
-
-
-
-
Summer Intern at Quantitative Research Department
-
Jun 2017 - Sep 2017
• Strategy principle: Identify abnormal trading volume from the relationship between tick interval (shortest stocks trading time) and the sudden increasing stocks price. • Extracted stocks trading data in 2017 with extreme daily return (above 3%) from SSE 50 Index (Shanghai Stock Exchange) and conducted data cleaning/preprocessing in Python and Wind software. • Analyzed the tick interval return and trading volume pattern of each stock to identify the abnormal trading volume using Python. • Investigated the abnormal tick interval each stock large trading volume and SSE Index based on data of return prices, bid/ask prices, etc in Python. • Designed the high-frequency statistical arbitraging strategy based on same tick interval difference between abnormal large trading volume stocks return and SSE Index return to determine the buy/sell decisions. • Created report to visualize the strategy in Python. • Implemented the new trading strategy in production, which generated annual return of 30%. • Further improved the trading strategy by expanding the selected stock ranges and trading periods. Show less
-
-
-
Indiana University Bloomington
-
United States
-
Higher Education
-
700 & Above Employee
-
Teaching Assistant
-
Sep 2016 - Jan 2017
• Answered students’ questions every class including finite math, basic linear algebra matrix and calculus, marked papers for quizzes and exams • Held department office hours (4 hours/week), organized and supported group discussions, prepared solutions for assignments and recitation materials • Answered students’ questions every class including finite math, basic linear algebra matrix and calculus, marked papers for quizzes and exams • Held department office hours (4 hours/week), organized and supported group discussions, prepared solutions for assignments and recitation materials
-
-
-
CITIC Securities Company Limited
-
China
-
Investment Banking
-
700 & Above Employee
-
Summer Intern at Industrial Research Department
-
Jul 2016 - Sep 2016
• Conducted new energy industry research by collecting and analyzing information such as reform plans about coal supply, energy market data, national policies from different states, etc. • Designed tables to store the industry research data using Access database and applied data cleaning. • Created quarterly new energy industry report using Excel VBA and checked effectiveness of data. • Conducted new energy industry research by collecting and analyzing information such as reform plans about coal supply, energy market data, national policies from different states, etc. • Designed tables to store the industry research data using Access database and applied data cleaning. • Created quarterly new energy industry report using Excel VBA and checked effectiveness of data.
-
-
Education
-
Columbia University in the City of New York
Master of Arts - MA, Statistics -
Indiana University Bloomington
Bachelor's degree, Applied Mathematics